Industrial Organization Taught by Dr. Prof. Fang Qiyun Textbook: Industrial Organization: A...

Industrial Organization

Taught by Dr. Prof. Fang QiyunTextbook: Industrial Organization: A Strategic

ApproachWritten by Jeffrey Church & Roger Ware

Part 1. Foundations

• Introduction

• The welfare economics of market power

• Theory of the firm

Ch.1. Introduction• 3 questions in IO• 1. Why are markets organized or structured as they are?• 4 aspects of market structure• (1)Firm boundaries. What determines the extent of a firm’s activities

in production? In particular, what are the factors responsible for determining the extent to which a firm is vertically integrated?

• (2)Seller concentration. seller concentration is a measure of the number and size distribution of firms. Industrial organization attempts to identify the factors that influence or determine seller concentration.

• (3)Product differentiation. Product differentiation exists when product produced by different firms are not viewed as perfect substitutes by consumers. What are the factors responsible for the extent of product differentiation.

• (4)Conditions of entry. The conditions of entry refer to the ease with which new firms can enter a market.

3 questions in IO• 2. How does the manner in which markets are organized affect the way i

n which firms behave and market perform?• If products produced by different firms are not viewed as perfect substitutes b

y consumers, then there will be a role foe non-price competition. In fact price competition might play a secondary role to other competition, such as product characteristics, advertising, and research and development expenditure.

• A major area of research in industrial organization is concerned with the theory of oligopoly: pricing behavior in a market dominated by a few large firms.

• A different kind of structural issue concerns the various institutions and practices adopted by oligopolistic firms to affect the nature of competition.

• Adam Smith: competitive markets were desirable because they led to outcomes that are socially optimal. Under certain circumstances, competition, as if guided by an invisible hand, results in the socially optimal level of output being produced at minimum resource cost, and distributed it to those who value it the most. IO is concerned with the efficiency or market performance of markets whose structure is not that are : What will the efficiency properties of imperfectly competition markets be, not just in terms of output but also in terms of product variety, quality, selection, and innovation? Is there a role for government intervention in terms of regulation or competition policy? Can we identify combinations of market structure and firm behavior where market outcomes are socially undesirable and susceptible to improvement? What are the economic foundations for regulation and antitrust policy? Why are intellectual property rights-patents, copyrights, and trademarks –created and enforced by the government?

3 questions in IO• 3. How does the behavior of firms influence the structure or organizati

on of markets and the performance of markets?• The emphasis of the previous question was on the effect of market structure

on the conduct of firms. The emphasis here is on adopting a more dynamic perspective and recognizing the possibility of feedback effects from firm conduct to market structure. We might expect that strategies which firms adopt today are intended to change market structure and thus firm behavior tomorrow. It would seem in fact that many aspects of non-price competition, such as research and development, are specifically designed to alert market structure tomorrow. Clearly, the extent of product differentiation is not determined only by exogenous factors such as the preferences of consumers. Firms have some latitude to choose the characteristics, range, variety, and quality of products they sell.

• Two issues that have received a great deal of attention are:• The potential strategies firms can adopt to drive competitors out of business

in order to establish a monopoly position.• The strategies that monopolists and oligopolists can adopt to deter the entry

of new competitors. These kinds of strategies obviously make seller concentration and barriers to entry endogenous.

The demand for IO

• 2 characteristics of markets in manufactured goods• (1)defferentiated products• (2)a few relatively large suppliers• Firms face downward sloping demand curves• Small numbers of competitors or the preference of

consumers for a specific products bestows some degree of market power on firms, and the competition will be imperfect.

• Market power is the ability to profitably raise price above marginal cost.

• IO is the study of the creation, exercise, maintenance, and effects of market power.

Methodologies (1)

• The methodology of a discipline refers to the basic approach(es) commonly used in a discipline in the creation of knowledge. It is a guide for practitioners about how to go about answering a question or solving a problem.

• The traditional approach in IO is the S-C-P (structure-conduct-performance) paradigm. The orientation of SCP approach is primarily empirical: researchers in this tradition try to uncover empirical regularities across industries.

• The “new IO” has been more concerned with developing and testing explanations of firm conduct.

• Key distinguishing features of the new IO:• (1)emphasis on specific industry• (2)focus on developing models of firm behavior• (3)empirical work is based on well-founded models of firm behavior

Methodologies (2)

• The ability to develop good theoretical explanations of firm behavior is due to the developments in non-cooperative game theory in the 1970s. Non-cooperative game theory consists of tools that are used to model the behavior or choices of agents (individuals, firms, etc.) when the payoff (profit) of a choice depends on the choices of others.

• The focus of the new IO on the conduct of firms in imperfectly competitive markets involves determining the factors and strategies that provide firms with a competitive advantage. With its focus on the nature and form of rivalry in concentrated markets, much of IO is a theory of business strategy.

Methodologies (3)

• IO distinguishes between strategic and tactical decisions. Strategic decisions have long-run implications for market structure-the competitive environment faced by firms. Strategic decisions involve things like product characteristics and capacity. Tactical decisions determine the short-term actions firms take given the current environment. The tactical decisions of a firm are usually either its price or output. Strategic decisions matter because by determining the current environment of a firm, they affect its pricing or output decisions. The ability of strategic variables to affect tactical decisions arises because of commitment. Strategic decisions commit the firm to follow a pricing policy or production level-because they are in its best interests-and that commitment depends on the irreversibility of the strategic decisions.

Methodologies (4)

• Students of IO and strategic management are concerned with identifying strategies which create monopoly rents and allow firms to maintain them. Of particular interest is the ability of firms to engage in profitable entry deterrence. An entry barrier is a structural characteristic of a market that protects the market power of incumbents by making entry unprofitable. Profitable entry deterrence-preservation of market power and monopoly profits-by incumbents typically depends on these structural characteristics and the behavior of incumbents post-entry. Appropriate strategic choices can commit an incumbent firm to act aggressively post-entry and insure profitable entry deterrence. In essence, firms can make investments that create barriers to entry or magnify/raise the importance of existing barriers to entry.

Methodologies (5)

• Anti-trust laws and competition policy are concerned with the creation and maintenance of market power. The intent of competition policy is to prevent firms from creating, enhancing, or maintaining market power. The new IO, with its focus on strategic competition and firm conduct to acquire and maintain market power, provides the intellectual foundation for determining when and why firm behavior and business practices warrant antitrust examination and prohibition.

Main issues in IO

• Perfect competition, economics of market power (the defining characteristic of imperfect competitive markets), the welfare economics used to assess market performance.

• The theory of the firm.• Different aspects of monopoly: its sources; its

costs and benefits; pricing; quality choice.• Theory of oligopoly pricing and the game theory.• Strategic competition.• Anti-trust economics.• Regulatory economics.

Foundations（ 1）• A review of perfect competition

• An introduction to the economics of market power-the defining characteristic of imperfectly competitive markets

• A discussion of the welfare economics used to assess market performance.

Foundations（ 2）• The theory of the firm:• (1) a review of the traditional microeconomic conception of a firm

where we review and highlight the relevance of a number of important cost concepts such as sunk expenditures, economies of scale, and economies of scope.

• (2)an extended discussion of the economics of organization in the context of trying to explain the boundaries of a firm. If markets are such an efficient institution to organize transactions, why are not all transactions organized by markets? Why do firms exist? Why do firms ever opt to make rather than buy? And why is it never more efficient to always make rather than buy? What limits the size of firms? Can we identify a set of factors that are responsible for determining whether a transaction is organized within a firm or by markets and thereby determine the extent of vertical integration? The limits to firm size are closely related to the objective of firms. In microeconomics the assumed goal of firms is profit maximization. However, when firms are controlled by professional managers and not shareholders this assumption may not be tenable. We examine the validity of this assumption and the mechanisms, both internal and external, that help align the incentives of owners and managers and in doing so promote profit maximization.

Monopoly (1)• Different aspects of monopoly: its sources; its costs and benefits;

pricing; and quality choice.• A discussion of the source of market power, highlighting the

importance of entry barriers. • Also consider two factors which might limit the ability of a

monopolist to exercise its market power: (1)the effect of product durability; (2)the possibility of a competitive fringe.

• An extended discussion of the costs and benefits of monopoly.• Analysis of how a monopolist might exploit her position by widening

the scope of her behavior: • (1)the monopolist may not charge the same price per unit across all

units and consumers-price discrimination occurs when different consumers pay different prices or the per unit price per customer varies across units. Explore the profit and welfare implications of price discrimination.

Monopoly (2)• (2)explore the questions of information, advertising, and quality.

Search goods-products whose quality consumers can judge through prior knowledge or by inspection at the time of purchase. Experience good-the quality of experience good can only be ascertained by consumers ex post.

• There are 2 possibilities.• (1)when the monopolist can adjust quality over time, the monopolist

has an incentive to claim high quality and sell low quality-this gives rise to a problem of moral hazard. The introduction of warranties provides a commitment device for the manufacturer against this activity. To the extent that warranties are not effective, then repeat purchase by consumers may also create an incentive for the provision of high quality. There may also be a role for independent firms to perform quality tests and inform consumers of the results.

• (2)when the quality of a product is fixed, but only the monopolist knows the quality of its product before purchase. This leads to the problem of adverse selection. Monopolists whose products are of low quality will claim the opposite and as a consequence consumers will be appropriately skeptical of all high-quality claims. The strategies that a high-quality manufacturer can follow is that it can credibly communicate its quality to consumers, particularly via the role of advertising.

Oligopoly pricing(1)• An overview of the theory of oligopoly pricing: • (1)reviews the classic models of oligopoly pricing when products are

homogeneous-static models of oligopoly pricing-competition is limited to a single period.

• (a) the Cournot model assumes that firms compete over quantities. We consider the derivation of equilibrium, comparative static results, and welfare implications when the number of firms is fixed and when there is free entry.

• (b) the Bertrand model assumes that firms compete over prices. This gives rise to the Bertrand paradox: when products are homogeneous and firms have constant and equal marginal costs, the competitive result that price equals marginal cost arises even if there are only two firms in the industry. This result is not robust to the introduction of capacity constraints and differentiated products. The relative merits and usefulness of the Cournot and Bertrand models. One of the main results of both static models of imperfect competition is that the equilibrium outcome is not a collusive outcome: oligopoly prices and aggregate profits are lower than those of a monopolist.

Oligopoly pricing(2)

• (2) dynamic models of oligopoly-how dynamic competition (more than one period) makes it possible for oligopolists to sustain collusion or maintain a cartel and share in monopoly profits. The factors make collusion more or less sustainable. The idea of facilitating practices. Facilitating practices are a response by firms within an industry that increase the likelihood that collusion can be sustained.


• (3) oligopoly pricing in differentiated products markets-the 2 types of models used to analyze competition in differentiated products markets are monopolistic competition and address models.

• Models of monopolistic competition are used to determine whether market outcomes are characterized by the socially optimal number of differentiated products: are there too many brands of some product? Given that production is characterized by economies of scale, there is an implicit trade-off between costs of production and the benefits of more variety. Introducing another variety increases average costs of production, but this must be compared to the gain associated with an increase in variety.

Oligopoly pricing(4)• Address models of product differentiation begin with the assumption that

each product can be described completely by its location in product space. The distribution of the preferences of consumers is also in product space, where their address represents their most preferred product. These types of models have been used to analyze whether or not the best set of products is produced. Adding another product means a closer match between available products and the most preferred variety of some consumers. However, increasing the number of products decreases the output of each, and if there are economies of scale, average production costs will be increasing in the number of products. Three types of strategic behavior are associated with product differentiation. This behavior involves the use of product differentiation by incumbent firms to profitably deter the entry of competitors. (a) brand proliferation; (b) brand specification; and (c) brand preemption. Vertical product differentiation-competition over quality. In these address models, ceteris paribus, all consumers agree on which products are preferred-are of higher quality. However, consumers differ in their ability to pay (incomes) and hence the most preferred product for any individual depends on the set of available products, prices, and her income. These models are used to determine the range of quality available in the market and how the strategic choice of quality can relax price competition and deter entry.


• (4) the approaches used by economists to empirically identify market power and its determinants. Two conceptually distinct approaches are: (a) the S-C-P paradigm and (b) the new empirical IO.


• Game theory: provide an intuitive, conceptual introduction to the techniques used to study oligopoly behavior and strategic competition.

• (1)Simultaneous move game-static games.

• (2)sequential or dynamic games.

Strategic behavior(1)

• The distinction between short-run (tactical) decisions and long-run (strategic) decisions. Strategic decisions, in part, determine both the possible tactical decisions and the payoffs associated with the tactical choices. In the context of IO, the tactical decisions usually involve prices or output. The strategic variables include plant capacity, advertising, product selection, and R&D.

Strategic behavior(2)• Introduction to strategic behavior and the importance of commitment.• Define a strategic move and explain how it converts an idle threat into a

credible threat (commitment) by changing incentives and expectations. • Early work on strategic behavior emphasized so-called indirect effects.

A move or action by A is strategic if it changes B’s expectations of how A will behave, and as a result alters the behavior of B in a manner favorable to A. In IO such a strategic move is usually associated with sunk expenditures or binding contracts supported by a legal framework. If one firm can move first and incur sunk expenditures, its production incentives will change.

• The relationship between sunk expenditures, strategic moves, and commitments. These concepts are then used to provide a consistent game-theoretic interpretation of the classic oligopoly model of Stackelberg. Show how a firm can successfully increase its market share and profits if it can commit to its level of output prior to its rival’s response by sinking its costs of production. This model also provides a natural starting point for considering the issue of profitable strategic entry deterrence: under what circumstances is it possible and profitable for an incumbent firm to deter the entry of an equally efficient rival?

• The limit-price model.

Strategic behavior(3)• Modern theory of entry deterrence and the synthesis of the two

existing views.• How and when investment in capacity can provide the means for an

incumbent to deter entry by credibly committing it to behave aggressively if an entrant should enter, thus rendering entry unprofitable. This strategic approach emphasizes how the sunk expenditures of the incumbent provide it with a strategic advantage by reducing its economic costs.

• An alternative perspective is offered by the theory of contestability. The contestability of a market is determined by the magnitudes of sunk expenditures incurred upon entry by an entrant. When there are no sunk expenditures associated with entry, hit-and-run entry provides a means whereby competition in the market is replaced by potential competition. If there are sunk expenditures associated with entry, then entrants will be reluctant to enter if they anticipate that these expenditures will not be recovered.

Strategic behavior(4)• The theory of 2-stage games or strategic competition.• Generalize from the modern theory of entry deterrence to the

development of the full taxonomy of business strategies. This taxonomy provides a guide to understanding how firms can identify, capture, and protect rents.

• A wide range of strategies include learning by doing, tying, choice of managerial incentives, lease-or-sell decisions, direct distribution or use of independent retailers, and switching costs.

• Two areas of corporate strategy: (a) advertising: informative and persuasive advertising; the incentives and effects, as well as the social desirability, for both kinds of advertising.

• (b) the economics of R&D: the special nature of knowledge and the implications of that nature for its production, the relationship between market structure and innovative activity, the rationale for patents and the determination of the characteristics of an optimal patent, and the efficiency implications of patent races.

Issues in antitrust economics(1)• Public policy responses to the exercise of market power-issues in antitrust e

nforcement and regulation.• Market definition, highlighting the differences between economic markets an

d antitrust markets. Market definition in antitrust is a search for market power. Without market power, firm conduct will not raise efficiency concerns. Various techniques to define antitrust markets and identify market power in practice.

• The theory of strategic behavior-direct strategic effects. Direct strategic effects arise when the profits of a rival firm depend directly on actions or investments by the firm. Practices that cause a direct negative effect on the profits of rival firms are termed exclusionary. There are two types of exclusionary practices associated with strategic investments. These types of investment either raise the costs of rivals or reduce their revenues. The effectiveness and profitability of several specific types of behavior are considered. These include the foreclosure effects when a firm merges with an input supplier and withholds supply from its rivals, outbidding rivals for scarce inputs, raising industry-wide input prices, controlling access to complementary products, advertising, and control of compatability standards.

Issues in antitrust economics(2)• A second type of exclusionary: predatory pricing. Predatory pricing

involves a firm setting prices to induce the exit of rival firms. Its motivation is to reduce competition and increase its market power or become a monopolist. We identify the circumstances when predatory pricing will be a successful and profitable exclusionary strategy.

• Vertical restraints. Vertical restraints refers to contractual restrictions imposed by manufacturers on the retailers that comprise their distribution channels. The main vertical restraints: franchise fees, resale price maintenance, quantity forcing, exclusive territories, and exclusive dealing. It provides an economic analysis of why they are utilized and a determination of their impact on efficiency.

• Horizontal merges. It contains a discussion of the motivation and effects of merges. The modern analysis of merges suggests that the effects of a merge depend on the impact on and response of non-merging firms. An extended discussion of the antitrust treatment and analysis of merges.

Issues in regulatory economics

• An overview of regulatory economics.• Economic justifications for price and entry

regulation.• Optimal pricing in a natural monopoly.• A number of issues in regulation: (1)the

implications for optimal pricing when there are asymmetries of information between the firm and the regulator, (2)the practice of regulation, (3)entry by regulated firms into unregulated markets, (4)access pricing to essential facilities.

Ch.2 The welfare economics of market power

• Profit Maximization

• Perfect Competition

• Efficiency

• Market Power

• Market Power and Public Policy

Profit Maximization• IO is about the behavior of firms in imperfectly competitive

markets.• To understand firm behavior we typically start by assuming

its objective is to maximize profits.• π(q)=R (q)-C (q)• MP (q)=MR (q)-MC (q)• MR (q) =MC (q)• Keep producing or shut down decision: In the short run, it i

s better to keep producing if price is greater than minimum average avoidable costs. In the long run, it is better to keep producing if price is greater than minimum long-run average costs. In the long run, all costs are avoidable. In the short run, some kinds of costs are not avoidable-sunk costs.

Perfect competition(1)• 4 standard assumptions of the perfectly competitive model:• Economies of scale are small relative to the size of the market. This means

that average costs will rise rapidly if a firm increases output beyond a relatively small amount. Consequently, in a perfectly competitive industry there will be a large number of sellers. We also assume that there are many buyers, each of whom demands only a small percentage of total demand.

• Output is homogeneous. That is, consumers cannot distinguish between products produced by different firms.

• Information is perfect. All firms are fully informed about their production possibilities and consumers are fully aware of their alternatives.

• There are no entry and exit barriers. This means that the number of firms in the industry adjusts over time so that all firms earn zero economic profits or a competitive rate of return. Because positive and negative economic profits create incentives for the number of firms in the industry to change. If economic profits are positive then the revenue of a firm exceeds the opportunity cost of its factors of production-the value of the inputs in their next best alternative use. Without entry barriers, entrepreneurs have an incentive to enter by transferring factors of production from other industries or activities. And without exit barriers, negative economic profits mean that firms will exit since their factors of production can, and will, be profitably transferred to other industries.

• Assumptions 1-3 imply price-taking behavior. Price takers believe or act as if they can sell or buy as much or as little as they want without affecting the price. In effect they act as if prices are independent of their behavior.

Perfect competition(2)• Supply• A single firm• R(q)=pq MR(q)=p p=MC(qc）• The relationship between price and the profit-maximizing output-the level of outp

ut that the firm would like to sell-is called the firm’s supply function. The supply function qc=S(p) is found by solving p=MC(qc) for qc. The inverse supply function is p=MC(qc).

• The firm’s long- and short-run shutdown decision:• It is better to stay in business if total revenues exceed avoidable costs, and • Sunk costs must be paid whether the firm stays in business or not. This means t

hat truly sunk costs are irrelevant to the shutdown decision. • Not all fixed costs are sunk-they also include quasi-fixed costs. A firm is better of

f producing where price equals marginal cost only if price is greater than average avoidable cost. In the long run all costs are avoidable, so a firm should stay in business only if price is greater than minimum average total cost. This means that regardless of the run, long or short, the firm’s supply curve is the relevant marginal cost curve above minimum average avoidable cost. Changing the time horizon changes the firm’s avoidable costs and marginal cost curve.

• The difference between total revenues and avoidable costs in the short-run equals the firm’s quasi-rents. Quasi-rents measures the benefit to the firm of staying in business. They are the difference between its revenues from staying in business and what is required for the firm to stay in business, its avoidable costs. Quasi-rents provide a contribution towards the firm’s sunk costs. In the long-run, all costs are avoidable, and the difference between total revenues and total costs is economic profit.

Perfect competition(3)

• Market supply • Market supply is the total amount firms in the in

dustry would like to sell at the prevailing price. The market supply function can be found by summing up the individual supply functions for each firm.

• S(p)=∑Si(p)

Perfect competition(4)• Market equilibrium• At the equilibrium price both firms and consumers are able

to fulfill their planned or desired trades: firms are able to sell their profit-maximizing quantities and consumers are able to purchase their utility-maximizing quantities. So the equilibrium price is the price that equates the quantity supplied with the quantity demanded: Qs(pc)=Qd(pc)

• Price-taking firms have firm demand curves that are horizontal and equal to the market price.

• The long-run competitive equilibrium price requires (a) that quantity demanded equal quantity supplied; and (b) that the number of firms adjust to economic profits are zero. The long-run equilibrium price must be equal to the minimum long-run average cost of production. Otherwise firms could adjust the scale of their operations and earn positive economic profits.

0 q

p

0 q

p

(a) firm (b) market

Competitive equilibrium

D

S

pc

pminpmin

pc

Qcqminqcqmin

AC(qc)

MC(q)

AC(q)Quasi-rents

Efficiency (1)

• Adam Smith first conjectured that competitive markets were desirable because the outcomes associated with them were socially optimal. It was as if an invisible hand was at work guiding the interaction between firms and consumers such that the socially optimal amount of output is produced at minimum resource cost and this output is distributed to those who value it the most. The key to Smith’s insight is understanding the idea that voluntary trade allows individuals to realize gains from trade and that as long as some gains from trade remain unexploited, there is an incentive for more trade.

Efficiency (2)• Measures of gains from trade• Consumer surplus• Consumer surplus is the answer to the question, how much would a

consumer have to be paid to forgo the opportunity to purchase as much as she wants of a good at a given price. It is the difference between the consumer’s willingness to pay for another unit of output and the price actually paid. The willingness to pay (WTP) for a unit of output is the maximum amount of money that the individual is willing to forgo in order to consume that unit of output. It is a dollar measure of the consumption benefit provided by that unit of output. If WTP>P, the consumer realizes gains from trade: the benefit from consuming the unit exceeds how much she has to pay. The difference between WTP and the actual price paid is the consumer surplus for that unit: WTP-P. The optimal consumption level is where the willingness to pay for another unit equals price. On the last unit consumed, consumer surplus is zero. Consumer surplus for an individual from total consumption is a dollar measure of the consumer’s gain from trading money for the good.

• A consumer’s demand curve shows her willingness to pay for each unit of output.

• Aggregate consumer surplus is a measure of gains from trade accruing to all consumers in a market, so it is simply the sum of the individual consumer surpluses. It is the area below demand curve and above the price line.

Efficiency (3)• Producer surplus• Producer surplus is the answer to the question, how much would a producer hav

e to be compensated in order to forgo the opportunity to sell as much as she wants at a given price? The benefit to a producer from producing in the short run is given by her quasi-rents. Quasi-rents provide a quantitative measure of how much better off producers are from trading. In the context of the gains from trade, quasi-rents are often called producer surplus. A firm’s quasi-rents are the difference between its revenues and total avoidable costs.

• The firm’s supply curve is its marginal cost curve above minimum average avoidable cost. A firm will prefer to shut down if P<Pmin where Pmin is the price at which the firm finds it optimal to produce where average avoidable cost is at a minimum (output level qmin). When P=Pmin a profit-maximizing firm breaks even. For P>Pmin, quasi-rent per unit for

• q>qmin is the difference between price and marginal cost.• q<qmin is the difference between price and Pmin.• An alternative way to think about the derivation of quasi-rents is that the benefit t

o a producer is the difference between what she receives (price) and what she has to pay to supply that unit.

• The difference between price and the minimum required for supply is the quasi-rent on that unit of output.

• If we sum up per-unit quasi-rent we get total quasi-rents.• Producer surplus is the area below the price line and above the supply curve.

q

P

0q

P

Pc

0

D

Consumer surplusProducer surplus

MC(q)

Pmin

qmin

Consumer and producer surplus

Efficiency (4)

• Total surplus• Total surplus is simply the sum of consumer and

producer surplus for a given quantity. On a per-unit basis it is the difference between consumers’ WTP and the minimum required for it to be supplied by producers. The minimum required for supply is marginal cost.

• The quantity of output that maximizes total surplus is where WTP=MC: at this level of output the amount of other goods consumers are willing to give up for one more unit exactly equals the amount of other goods they have to give up.

Q

P

S

D

Qc

Pc

0

Aggregate consumer surplus

Aggregate producer surplus

Gains from trade in competitive equilibrium

Total surplus=aggregate consumer surplus + aggregate producer surplus

Efficiency (5)• Pareto optimality• An outcome is Pareto optimal if it is not possible to make one person

better off without making another worse off.• A move from allocation or outcome A to B that makes someone

better off-a winner-without making someone else worse off-a loser-is a Pareto improvement (PI).

• A move from A to B is a potential Pareto improvement (PPI) if the winners could compensate the losers and still be better off, but they don’t. If compensation is paid, the change is no longer potential-it is an actual PI.

• Adoption of the PPI criterion means that we can focus on what happens to total surplus. Using the PPI criterion amounts to asking if a change increases the size of the pie, without asking about the distribution of the pie.

• An outcome or allocation for which total surplus is maximized implies that there are no unexploited gains from trade available and therefore it is Pareto optimal.

• A Pareto optimal state is efficient.

Efficiency (6)• There are 3 well-known problems with assessing efficiency on the basis of

changes in total surplus:• 1. Consumer surplus is not an exact monetary measure of consumer welfare. It

is, however, a good approximation to the two exact measures (compensating variation and equivalent variation) if the income effect is small. The income effect measures the effect of price changes on income and the effect of those income changes on demand for the good. Changes in consumer surplus are a good approximation if demand is not affected much by the income effects of a price change and not a good approximation when demand is affected significantly by the income effects of a price change.

• 2. The basis of consumer and producer surplus is that demand and supply curves represent not only private benefits and private costs (which they clearly do), but also capture all social costs and benefits as well. This will not be the case if there are externalities. Negative externalities means that the total amount of other goods forgone is greater than that represented by the supply curve, or that the amount consumers in aggregate are willing to give up to increase consumption of the good is less than that represented by the demand curve. If a positive externalities exists, the conclusion is reversed.

• According to the theory of the second best, maximization of total surplus in one market may not be efficient if surplus in other markets is not also maximized.

• 3. Distribution of the gains from trade is not explicitly taken into account when changes in total surplus are used to rank outcomes. There is an implicit assumption that a dollar of consumer surplus is identical in value to society as a dollar of producer surplus. This may not be universally accepted.

Market power

• A firm has market power if it finds it profitable to raise price above marginal cost. The ability of a firm to profitably raise price above marginal cost depends on the extent to which consumers can substitute to other suppliers. It is possible to distinguish between supply and demand substitution. Supply side substitution is relevant when products are homogeneous, whereas demand side substitution is applicable when products are differentiated.

Supply substitution

• The potential for supply substitution depends on the extent to which consumers can switch to other suppliers of the same product. If consumers cannot substitute to other suppliers capable of making up all or most of the reduction in its output, a producer of a homogeneous good will have market power.

Demand substitution

• The potential for demand substitution depends on the extent to which other products are acceptable substitutes. If products are sufficiently differentiated so that they are not close substitutes, then some consumers will not substitute to other products when price raise above marginal cost.

• A firm with market power is often called a price maker. A price maker realizes that its output decision will affect the price it receives. The demand curve that a price-making firm faces is downward sloping.

Market power and pricing• A firm is a monopolist if it believes that it is not in competition with ot

her firms. A monopolist does not worry about how and whether other firms will respond to its prices. Its profit depend only on the behavior of consumers (as summarized by the demand function), its cost function (which accounts for technology and the prices of inputs), and its price or output. A firm will be a monopolist if there are no close substitutes for its product. This means that the cross-price elasticities of demand between the product of the monopolist and other products are small (and vice versa).

• Εij=∆qi/∆pj• If the cross-price elasticities between the monopolist and other firms

are small, then changes in the price charged by the monopolist will have very little effect on the demand for the products supplied by other firms. Hence it is unlikely that they will respond. Moreover, if the cross-price elasticity between the other firms and the monopolist is small the effect of any response on the demand for the monopolist’s product will be sufficiently trival that it can be ignored by the monopolist.

Monopoly pricing

• The profit function of the monopolist is• Π(Q)=P(Q)Q-C(Q)• The profit-maximizing output level is defined by• MR(Q)=MC(Q)

P

QQ1 Q1+1

P(Q1)

P(Q1+1)P=P(Q)

Gains on marginal unit

Loss on inframarginal units

Marginal revenue of a monopolist

Inefficiency of monopoly pricing (1)

• The socially optimal quantity is found where marginal cost equals the marginal benefit of consumption. Monopoly pricing affects both the magnitude of gains from trade and their distribution. Monopoly pricing is inefficient since the monopolist produces two few units.

• The difference between the total surplus under monopoly and maximum total surplus is called deadweight loss (DWL). It represents an opportunity cost to society.

• A second effect of monopoly power is the transfer of surplus from consumer to the firm as profits. Under competitive pricing, both monopoly profit and the DWL would have gone to consumers as surplus. In order to realize a large share of the gains from trade, the monopolist raise price above marginal cost. This comes at a cost to society in the form of lost surplus, since some consumers respond to the price rise by reducing their quantity demanded.

Inefficiency of monopoly pricing (2)

• Since total surplus is not maximized by a monopolist, potential Pareto improvements must be possible. There are many ways in which the gains from trade represented by DWL could be realized. For instance, consumer could band together and form a society. The society could response to the monopolist that it set its price equal to marginal cost, and in return the society would pay a lump sum equal to πm+t, where t is small. As a result the profit of the monopolist would increase by t and the surplus of consumers by DWL-t. The problem with this scheme is that the costs associated with organizing consumers are likely to be large.

P

Qs

Pm

QQm0

MC=c

P=P(Q)MR(Q)

DWL

Consumer surplus

Monopoly profit

Profit-maximizing monopolist

Example: monopoly pricing with constant marginal costs and linear demand

• Suppose that (1) P(Q)=A-bQ and MC(Q)=c. Find the monopoly price and output.

• Solution• MR(Q)=A-2bQ=MC(Q)=c• Qm=(A-c)/2b• Pm=(A+c)/2 if A>c then Pm>c• Πm=(A-c)2/4b• DWL=(Pm-c)(Qs-Qm)/2• Qs=(A-c)/b• DWL=(A-c)2/8b• CS=(A-Pm)Qm/2• CS=(A-c)2/8b

Measurement and determinants of market power (1)

• MR(Qm)=P(Qm)+dP(Qm)/dQQm=MC(Qm)• P(Qm)[1+dP(Qm)/dQQm/Pm]=MC(Qm)• P(Qm)[1-ε]=MC(Qm)• ε = -dP(Qm)/dQQm/Pm• L=[ P(Qm)- MC(Qm) ]/ P(Qm)=1/ ε • The Lerner index L is defined as the ratio of the firm’s profit margin P(Qm)-

MC(Qm) and its price. It is a measure of market power since it is increasing in the price distortion between price and marginal cost. It shows that the market power of a firm depends on the elasticity of demand ε .

• In considering a monopolist, we did not have to distinguish between the market demand curve and demand curve of the firm-they were the same. However, in general a firm may have market power and not be a monopolist. The extent to which a firm in imperfectly competitive markets can exercise market power depends on the elasticity of its demand curve. The greater the number of competitors (for homogeneous goods) or the larger the cross-elasticity of demand with the products of other producers (for differentiated products), the greater the elasticity of the firm’s demand curve and the less its market power.


• The extent of the inefficiency associated with market power also depends on the time frame. In the long run, a firm’s elasticity of demand is likely to be larger for 3 reasons:

• (1) Consumer response: long run vs. short run. The long-run response of consumers to a price increase is often greater than their short-run response.

• (2) New entrants. If economic profit are positive, then other firms may try to enter the market. Entry of any magnitude increases the elasticity of the firm’s perceived demand curve, reducing its market power. A monopolist may even become a price taker if entry is sufficiently extensive.

• (3) New technology. Technological change can generate new products and services, and the introduction of these products reduces the market power of producers of established products. In some cases entire industries are virtually wiped out by the effects of technological change: consider the fate of typewriters.


• The last 2 factors suggest that the ability of a firm to exercise market power in the long run will depend on barriers to entry. If entry is easy, then we would not expect firms to have significant market power in the long run.

• Entry and competition from other products (demand side substitution) and other producers (supply side substitution) will limit, if not eliminate, a firm’s market power if entry barriers are insignificant. On the other hand, if entry barriers are significant, then a firm will be able to exercise market power even in the long run.

The determinants of DWL• DWL does not vary inversely with the elasticity of demand. The size of

the DWL depends on both the Lerner index (which varies inversely with the elasticity of demand) and the quantity distortion, the difference between Qs and Qm (which varies directly with the elasticity of demand). When demand is less elastic, the price distortion is large, but the efficiency implications of this are partially offset by the fact that the quantity distortion will be less. This means that when demand is relatively less elastic, the transfer of surplus associated with monopoly pricing is large, but the inefficiency or DWL is small.

• The DWL associated with monopoly pricing is approximately equal to DWL=dPdQ/2.

• If we assume constant cost MC(Q)=c, so that dP= Pm-c, then• DWL=(dQ/Q)/(dP/P)PQ(dP /P)2/2=εPQL2/2, L=dP/P• This suggests that the inefficiency associated with monopoly pricing is

greater, the larger the elasticity of demand ε, the larger the Lerner index L, and the larger the industry (as measured by the firm’s revenues PQ). However, such an interpretation would be incorrect since L depends on the elasticity of demand ε. As εincreases, a profit-maximizing monopolist responds by decreasing L.

Market power and public policy• Public policy towards market power takes one of two forms. Concerns regarding the

inefficiency associated with the exercise of market power typically result in regulation. Regulation involves government intervention to limit the exercise of market power, typically by constraining or limiting prices. Antitrust laws, on the other hand, are suppose to limit the acquisition, protection, and extension of market power. They do so by making certain kinds of behavior illegal. The economic rationale for determining the legality of behavior is to assess its effect on market power. Behavior that creates, maintains, or enhance market power should be prohibited because of the DWL from the exercise of market power.

• In the economic approach to determining the legality of a firm’s behavior we ask: what are its effects on total surplus? If total surplus declines, the behavior should be illegal. If total surplus increases, then there is a presumption on economic grounds that the behavior is desirable and should be legal.

• Consider, for instance, the legality of agreements to fix prices. In the United States, courts have distinguished between naked restraints and ancillary restraints. A price-fixing agreement is deemed a naked restraint if the objective and effect of the agreement are to restrict competition. Naked restraints are per se illegal. If firms agree to fix prices, the agreement is illegal, regardless of the firm’s intentions or the economic effects of the agreement. The reasoning is based on the believe that firms enter into such price-fixing agreements to curtail competition, increase their market power, and charge monopoly prices.

• Ancillary agreement, on the other hand, are agreements whose primary purpose and effect are not to fix prices, but to achieve some other legitimate business objectives. That is, the fixing of prices is not the main purpose, but attaining the objective of the agreement requires fixing prices. In these cases, the legality of a price-fixing agreement is subject to a rule of reason approach. Under a rule of reason approach it is recognized that certain aspects of the behavior might be welfare improving, but for the agreement not to be an unreasonable restraint of trade, these aspects must be sufficient to offset the inefficiency associated with the presumed increase in market power.

Summary • Profit-maximizing firms produce where their marginal revenue equals

marginal cost.• If markets are perfectly competitive, the allocation of resources is

Pareto optimal or efficient. An efficient allocation maximizes total surplus.

• A firm with market power can profitably raise price above marginal cost. The exercise of market power creates an opportunity cost to society called deadweight loss (DWL). In raising price above marginal cost, units of output for which the value to consumers exceeds marginal cost are not produced.

• The market power of a firm varies inversely with its elasticity of demand. Supply side (other producers of the same product) and demand side substitution (competing products) possibilities for consumers increase the elasticity of demand. Barriers to entry determine the extent to which a firm can exercise market power in the long run.

• Deadweight losses provide an economic rationale for state intervention. Regulation is intervention to constrain the exercise of market power, while antitrust laws make behavior that creates, extends, or preserves market power illegal.

Ch.3 Theory of the Firm

• Neoclassical theory of the firm

• Why do firms exists?

• Limits to firm size

• Do firms profit maximize?

• Summary

Neoclassical theory of the firm• The traditional approach in microeconomics is to define a firm by its pr

oductive activities. A firm is defined by a set of fisible production plans completely described by a production function. The production function maps bundles of inputs into output. The firm-or implicitly its managers-determine how, what, and how much to produce. The assumed objective is profit maximization, which incorporates cost minimization.

• The cost function summarizes the economically relevant production possibilities of the firm. The cost function C(q) gives the minimum cost of producing q units of output. It incorporates both technological efficiency and the opportunity cost of inputs. Technological efficiency means that the firm uses no more inputs than necessary to produce q. and of all those input bundles that are just able to produce q, the firm chooses the one with the minimum opportunity cost.

• The average cost function of a firm is the minimum cost per unit produced: AC(q)=C(q)/q

• The marginal cost of production is the increase in total costs if output is increased marginally. It is the rate of change in total cost wity respect to output: MC(q)=dC(q)/dq

• At theleval of output for which average cost is minimized, MC=AC.

q

c

0

AC(q)MC(q)

qmes

Average and marginal cost functions

Review of cost concepts (1)• 1. Opportunity cost• 2. The economic costs of durable inputs• In general the opportunity cost (OC) of using a durable asset in period t is OC

=Pt-Pt+1+iPt

• The user cost of capital r is found by dividing through by the initial value of the asset Pt.

• r=δ+i• δ=(Pt-Pt+1)/Pt

• 3. Avoidable costs and sunk expenditures• An avoidable cost is a cost that can be avoided by not producing. In contrast

a sunk expenditure cannot be avoided if the firm stops producing. Sunk expenditures arise because productive activities often require specialized assets. Specialized or specific assets cannot easily be used in other productive activities. The portion of an expenditure that is sunk is the difference between its ex ante opportunity cost and its salvage value or opportunity cost ex post. It is the portion of costs that are not recoverable upon exit from the original productive activities.

• We can distinguish between industry- and firm-specific capital. Example: airplane to the airline industry and an airline.

Review of cost concepts (2)• 4. The short run versus the long run• Economists typically talk about the short run as the period in which some

factors are fixed and the long run as the minimum time period such that all factors are in variable supply. In reality all factors can always be varied to some extent, but there are two constraints on how quickly a production process can be changed to a new arrangement or the utilization of some inputs changed. (1) the avoidable costs of the existing production process do not include sunk expenditures, but the avoidable costs associated with a new production process include all costs. In particular, additional investments in factors that ex post are sunk are costs ex ante. (2) time is required to make the investments associated with a change to a different production process or to adjust the utilization of some factors of production. The speed at which utilization of factors of production is made determines the cost of adjustment.

• 5. Variable and fixed costs• Variable costs vary with the rate of production. Fixed costs do not. Variable

costs are avoidable. In the short run fixed costs are either avoidable (quasi-fixed) or sunk. Short run and long run fixed costs.

The potential advantages of being large

• Large firms can have lower per unit costs than smaller ones, but we must be careful in distinguishing several different scale effects and the reasons for their existence. It is useful to differentiate between the advantages of being large at the product level (economies of scale), the plant level (economies of scope), and the level of the firm (multiplant economies of scope).

Economies of scale (1)• Potential per unit cost advantages from producing more of the same

product arise from economies of scale. Economies of scale exists if long-run average cost declines as the rate of output increases. If long-run average cost increases/stays constant when output increases, the technology is characterized by diseconomies of scale/constant returns to scale.

• Since average cost is falling/rising when it exceeds/is less than marginal costs, we can define a measure of economies of scale S as

• S(q)=AC(q)/MC(q)• S(q)>1 indicates that there are economies of scale at that output level.

Economies of scale are global if S(q)>1 for all levels of output. The rate of output where average cost is minimized and economies of scale are exhausted is called minimum optimum scale (MOS) or minimum efficient scale (MES).

• The concept of economies of scale-which is based on the behavior of costs-is closely related to the idea of returns to scale-which is based on technology.

Economies of scale (2)• Economies of scale arise because of indivisibilities. Indivisibilities

arise when it is not possible to scale some inputs down proportionately with output. Indivisibilities mean that it is possible to do things on a large scale that cannot be done on a small scale.

• The following are examples of indivisibilities that create economies of scale:

• 1. Long-run fixed costs• An input is indivisible if there is some minimum size below which it

becomes useless or does not exist.• An indivisible input can produce over some range of output before its

capacity is reached. Over this range there will be economies of scale: output can be expanded without increasing the amount of indivisible input. The cost of the minimum size input required for production is a long-run fixed cost. Spreading long-run fixed costs over a larger output reduces per unit fixed costs, leading to decreasing average costs over at least some range of output. Marketing and advertising costs are often fixed costs that contribute to economies of scale.

Economies of scale (3)

• 2. Setup costs• Before a firm can begin producing it is often the case

that it must first incur fixed setup or startup costs. These costs are incurred prior to production and do not vary proportionately with production. Indeed they are often invariant to the level of output. As a result, the larger the volume over which the setup costs are spread, the lower will be average costs.

• An important class of setup costs in some industries are research and development (R&D). The purpose of R&D efforts is to create new products, improve existing products, improve existing production processes, and/or develop new production processes.


• 3. Specialized resources and the division of labor• Adam Smith pointed out over 200 years ago that

specializing tasks through the division of labor resulted in an increase in productivity and therefore lower unit costs. Smith attributed that the increase in productivity to three factors: (1) increased dexterity or skill of workers; (2) the savings in setup costs; and (3) the substitution of specialized machinery for skilled craftsmen.

• The same principle is applicable to specialized capital. If output is lowered and capital is indivisible you cannot use just a proportion of the machine. Instead, cost-minimizing firms typically substitute a different type of machine that is not quite as efficient at larger rates of output, but is more efficient at smaller rates of output.


• 4. Volumetric returns to scale• Volumetric returns to scale or dimensional

economies can occur in any product or process involving containers. Capacity or output depends on volume, but the costs of the container depend on its surface area.

• 5. Economies of massed reserves• At low levels of output it may be necessary to

have relatively large inventories of replacement parts and backup machinery. However, as output increases, the ratio of the reserves to operating equipment can fall.

Economies of scope (1)• The cost efficiencies of being large at the plant level arise from econo

mies of scope from producing more than one product. Economies of scope in the two-good case exist if costs satisfy the following inequality:

• C(q1,q2)<C(q1,0)+C(0, q2)• Economies of scope exist if it is cheaper to produce the two output le

vels together in one plant than to produce similar amounts of each good in single-product plants.

• Economies of scope are also attributable to indivisibilities. The most common case occurs when facilities and equipment are indivisible, but not so highly specialized that they can only be used to produce one product. They are shared indivisible inputs. In these instances if the capacity of the indivisible input exceeds the firm’s production requirements, it can use that capacity to produce other products. The existence of common or shared factors is a compelling explanation for the existence of multiproduct firms. One view of a firm is that it is not in business to sell its output, but to sell its capacity. It will produce whatever products it can in order to maximize its capacity utilization.

Economies of scope (2)• The cost of the shared or common input is common to the set of prod

ucts or services that it produces. A cost is common if once incurred to produce product A, the cost does not have to be reincurred when product B is also produced. Alternatively, common costs are not attributable to any individual product. Attributable costs of a product are its incremental costs. Incremental costs for a product equal the difference between total costs with the product and total costs without the product, holding the production of all other outputs constant. The common costs of a firm are the difference between its total costs and the sum of the incremental costs for each product. The larger common costs as a proportion of total costs, the more important economies of joint production. Common indivisible inputs can give rise to fixed common costs.

• Economies of scope also exist if production involves a pure public input. Such an input is acquired to produce one product, but can then be costlessly used in the production of other products. Such an input does not become congested when used to produce a single product. A pure public input underlies examples of joint production. Joint production occurs when products are produced in fixed proportion.

Multiplant economies of scope

• Multiplant economies of scope arise from inputs that are indivisible at the level of the firm. These inputs can be shared across plants and products. Examples include specialized inputs, commonly known as corporate overhead, such as strategic planning, accounting, marketing, finance, and inhouse legal counsel. Two other important examples are distribution channels and knowledge.

Economies of scale and seller concentration (1)

• Consider how economies of scale interact with demand to provide a cost-based theory of seller concentration when products are homogeneous. If the minimum efficient scale (MES) is large relative to the quantity demanded, there will not be room for many cost-efficient firms. If the competition among firms results in prices that reflect minimum or efficient unit costs c*, then only a handful of firms can coexist when there are extensive economies of scale.

• 4 possible cases• 1.Constant returns to scale• For constant returns to scale there is no advantages or disadvantages

to being either small or big. In this case we cannot say very much about seller concentration, but we can say something about the equilibrium price. A market price above long-run average cost will result in incumbent firms earning positive economic profits. Since there are no disadvantages to producing at a small scale, this should invite entry, profits will eventually be competed away, and price will fall to long-run average cost. Only if price equals long-run average cost c* will there not be an incentive either for entry or for an incumbent firm to expand. If the price exceeds c*, firm will have an incentive to expand or enter. If the price less than c*, firms will contract or exit.

Economies of scale and seller concentration (2)• 2.Diseconomies of scale• In this case there is a cost disadvantage to producing more than one unit of

output. Efficient production requires many small firms, each producing one unit of output. In fact it is hard to see why firms would exist in this case: this case corresponds to household production. Each consumer produces her own requirements, and firms, as we usually think of them, and a market do not exist.

• 3.Economies of scale• When there are economies of scale there are obvious cost advantages to

being large. Indeed to minimize production costs a single firm is efficient. If the cost disadvantage associated with being relatively small is significant, then the market is likely to be dominated by a few large firms. Economies of scale mean that marginal cost is less than average cost. If increases in the number of firms imply that prices are more likely to reflect marginal costs, price-marginal cost margins sufficient for firms to earn normal profits (break even) require limits on the number of firms-that is, a lower bound on concentration. If the industry is initially characterized by concentration less than this minimum bound then prices will not be at a level that allows firms to recover their average costs. In the long run concentration will increase-through exit or merger and consolidation-until price-marginal cost margins are sufficient for firms to at least break even. When there are economies of scale the exercise of market power is necessary for a viable industry: market power is created by reducing the number of firms and increasing the size of those that survive.

Economies of scale and seller concentration (2)

• 4.U-shaped cost curve• In the case of U-shaped cost curves the equilibrium mark

et structure depends on the relationship between the MES and the size of the market. If the MES is small relative to the level of demand, then the market structure is likely to be similar to perfect competition, with many firms competing and price in equilibrium being driven to minimum average costs. Since some economies of scale are present, we do expect to observe firms of nonnegligible size. If the market is not large relative to the MES, then only a few firms can remain viable. The conditions necessary for perfect competition are no longer present, and we expect to see some form of oligopolistic competition, if not monopoly.

0q

AC

0q

0q 0

q

AC

AC AC

AC(q)=c*

c*

AC(q)

AC(q)

c*

AC(q)

c*

qmes

Constant returns to scale Diseconomies of scale

Economies of scale U-shaped cost curve

Seller concentration and economies of scale

Why do firms exist

• In traditional microeconomics the existence of the firms is taken as given. The organization and activities of a firm are assumed to be described by a production function and the objective of the firm is to maximize profits. The traditional approach, however, does not in fact offer an explanation for either the existence or limits on the size of firms. When we talk about the size of the firm-and its limits- there are two dimensions. The vertical scope of the firm refers to the number of stages in the vertical chain of production undertaken by a firm. The horizontal scope of the firm refers to how much of any given product it produces. The traditional technological view of the firm as a production function does not provide explanations for either the vertical or horizontal scope of a firm.

Two puzzles regarding the scope of a firm (1)

• Diseconomies of scale• Diseconomies of scale would seem to imply that the optimal size or

horizontal boundary of a firm is minimum efficient scale. Beyond this level unit costs start to increase. However, what are the sources of diseconomies of scale? The usual explanation is that some factors cannot in fact be replicated, meaning that diseconomies of scale arise not from variations in scale, but from the inability to in fact vary all factors: diseconomies of scale arise from factor substitution. The factor usually identified as a common source of diseconomies of scale is management. Management is thought to be a fixed factor that cannot be replicated. However, the theory does not explain why a second manager or management team cannot be hired to operate a second plant. Because it is silent on why firms cannot expand horizontally, the traditional view of the firm is more accurately characterized as a theory of plant size, not horizontal firm size.

Two puzzles regarding the scope of a firm (2)• Vertical boundaries• The vertical boundaries of a firm are determined by the number of

stages of the vertical chain of production it performs itself and which intermediate products it purchases from other firms. They are determined by what it decides to make and buy. The 5 main stages in the process of converting raw materials into goods available for sale to consumers are: (1) raw materials; (2) parts; (3) systems (parts are assembled into systems); (4) assembly (systems are assembled into final goods); and (5) distribution to consumers. The stages of production are linked by transportation and storage (warehousing). The vertical chain of production also requires corporate overhead or support services. These include activities such as accounting, legal services, finance, and strategic planning.

• The interesting question is which of these stages and activities (support services and transportation and storage) will be done internally and which will be sourced in the market. The traditional microeconomic theory of the firm is silent regarding the distribution of these stages between the firm and outside suppliers. If anything the traditional view was that in order to take advantages of economies of scale (from specialization and the division of labor) all of these activities need to be coordinated within the firm. However, the problem with this view is that it does not explain why the transactions between the different stages could not be coordinated using the price system or market.

Raw materials

Parts

Systems

Assembly

Distribution

Support services:

Accounting

Legal services

Finance

Data processing

Information system

Strategic planning

Human resources

Transportation

And

warehousing

The vertical chain of production: from raw materials to final goods

Explanations for the existence of firms• Why do firms exist?• As Coase noted, a little thought indicates that the existence of firms i

s in fact a puzzle. According to Coase, one of the hallmarks of what constitutes a firm, if not its defining criterion, is that production is organized by command. When production occurs within a firm, quantities produced are determined not by markets, but instead by overt and explicit coordination by management. This conscious suppression of the price mechanism is a puzzle since the use of prices and market exchange to direct and coordinate production is typically assumed to result in both cost minimization and exhaustion of gains from trade-both allocative and cost efficiency. If markets are so effective, why are there firms? Why do so many firms organize so many transactions or activities internally when they could use independent suppliers in the market?

• What determines the size of firms? Given that firms exist, which presumably means there are advantages to organizing production within a firm, why is not all production organized within a single firm? What factors limit the relative advantage of internal organization over market transactions, thereby bounding the size of firms? The answers to these two questions provide insight into the factors that determine the boundaries of a firm-what activities are organized within a firm and what activities are organized by the market. The answers determine both the horizontal and vertical scope of the firm.

Alternative economic organizations

• 3 alternative organizations or governance alternatives:• 1. Spot market: buy input B in spot market• 2. Long-term contracts: contract with supplier of input B• 3. Vertical integration: produce input B internally• The choice of governance alternative for a transaction depends on

its relative efficiency in adapting the terms of trade as conditions change.

Raw materials Input B Product A Customers

Make or buy

Spot markets (1)• The advantage of using spot market to source input B are threefold:

(1) efficient adaptation; (2) cost minimization; and (3) realization of economies of scale.

• 1. Efficient adaptation• The world is not static. Market conditions and opportunities are

dynamic and uncertain. Changes in demand and supply require adjustments in prices and quantities traded to realize all of the gains from trade. An advantage of relying on competitive spot markets for sourcing an input is efficient adaptation. Assume that input B is produced in a competitive market at constant unit cost. Then supply in the market will be perfectly elastic at price equals marginal cost. The use of the market results in efficient adaptation to changes in demand and cost. Equilibrium prices and quantities adjust to reflect changes in demand and cost and realize maximum total surplus.

QQ3BQ1

BQ2B0

P

P2B

P1B

MC2B

MC1B

MC3B

D3

D1

Spot market

Spot markets (2)• 2. Cost minimization• A residual claimant is the recipient of the net income from a project: they rec

eive whatever is left from an income stream after all other expenses have been deducted. This means that they internalize all of the marginal benefits from investments in cost reduction and/or efforts to reduce costs.

• Example: residual claimancy, high-powered incentives, and cost efficiency• Suppose that the profits of a price-taking input supplier are given byπ(q,e)=

pq-c(q,e)-e• Where the costs of production c(q,e) depend not only on the output level of t

he firm (q) but also its investment in cost reduction (or its effort to minimize costs) e. increases in e reduce the cost of the firm. The rate at which increases in effort reduce costs is given by dc/de<0. find the profit-maximizing effort and output.

• Solution: the profit-maximizing output for a price-taking firm (as always) equates price equal to marginal cost. The firm will invest in cost reduction until the marginal benefits of cost reduction equal the marginal cost:

• -dc(q*,e*)/de=1• Where q* and e* are the profit maximizing quantity and effort level.

Economies of scale

• The final advantage to using markets to source inputs is the potential for minimizing costs of production when there are economies of scale. If the demand for an input by a firm is less than minimum efficient scale, then by buying the input in the market it the market it might still be able to realize the cost advantages of production at minimum efficient scale.

qmesq1

pc=c*

AC1

AC(Q)

Economies of scale and outsourcing

Supplier switching

• The advantages of using spot markets, in particular, efficient adaptation and cost minimization, arise because there is no relationship between a firm and its input suppliers. The firm is indifferent between any suppliers, and the value of spot market arises because of the ability to costlessly switch suppliers. The firm can substitute away from suppliers that are high cost or are not willing to adjust quantities to maximize gains from trade. Incentives for integration must therefore arise only if there is something that locks firms to their suppliers so that they do not find it easy to switch. That something is relationship-specific investment.

Specific investment and quasi-rents

• In many instances in order to realize all of the potential gains from trade, both the firm and its output suppliers must make relationship-specific investments. The increase in gains from trade associated with relationship-specific assets arises from cost economies or tailoring design to the needs of a particular trading partner. Specificity of the investment to the trading relationship arises if the asset has limited value or use if the parties to the transaction change: either additional costs must be incurred or the productivity of the investment is reduced if it is redeployed to support exchange with another trading partner. In the extreme, an asset is specific only to trades between a firm and one input supplier. The cost of the investment is a sunk expenditure. The investment or some amount of it will not be recovered if there is a switch to another trading partner. The investment specific to the trading relationship locks in the supplier and the firm. The existence of relationship-specific investments means that an input supplier and a buyer will have an incentive to enter a long-term relationship.

Asset specificity• There are 4 common forms of asset specificity.• 1. Physical-asset specificity• Equipment and machinery that produce inputs specific to a particular customer

or are specialized to use an input of a particular supplier are examples of physical asset specificity.

• 2. Site specificity• Site specificity occurs when investments in productive assets are made in close

physical proximity to each other. Geographical proximity of assets for different stages of production reduce inventory, transportation, and sometimes processing costs. So called thermal economies are realized from the fuel savings since side-by-side location means it is not necessary to reheat the intermediate inputs. Specificity arise, however, because in many instances the assets are not likely mobile-they cannot be relocated at all or without incurring substantial cost.

• 3. Human-asset specificity• Human-asset specificity refers to the accumulation of knowledge and expertise

that is specific to one trading partner.• 4. Dedicated assets• Dedicated assets by an input supplier are investments in general capital to

meet the demands of a specific buyer. The assets are not specific to the buyer, except that if the specific customer decided not to purchase, the input supplier would have substantial excess capital.

Quasi-rents and the holdup problem• The quasi-rent associated with a specific investment is the difference

between the value of the asset in its present use-the ex ante terms of trade-and its next best alternative use, its opportunity cost. Quasi-rents provide a measure of the specificity of investment. The ex ante terms of trade provide sufficient incentives for the parties to agree to make the relationship specific investments and engage in trade. The opportunity cost of the investments ex post provides bounds on the terms of trade- after the relationship specific investments have been made-that make the trading partners willing to continue to trade and not exit or terminate the trading relationship.

• Relationship-specific investments imply a fundamental transformation. Suppose ex ante that there are many input suppliers and many buyers. This will not be the case ex post after relationship-specific investments have been made: alternative trading partners for both input suppliers and firms will be reduced. Ex ante there are many possible trading partners and competitive bidding is possible, but ex post the situation is characterized by small numbers and bargaining. The risk of opportunism-having your quasi-rents expropriated by an opportunistic trading partner-is illustrated by the following example.

The holdup problem• The risk of having your quasi-rents expropriated by an opportunistic trad

ing partner is called the holdup problem. The incentive exists for both sides to try and redistribute quasi-rents in their favor. The actual division in any instance will depend on the relative bargaining positions, abilities, and strengths of the trading partners. We would expect that parties that have relatively attractive alternatives-and thus whose loss from switching trading partners is less-will have stronger bargaining position. On the other hand, the more difficult it is to redeploy assets, the greater the quasi-rents of a firm and the more vulnerable it is to hold up.

• Masten (1996) has underlined the importance in some instances of temporal specificity. Temporal specificity arises when the timing of performance is critical. Masten identifies 4 situations where temporal specificity is likely to be important: (1) the value of a product depends on it being delivered in a timely manner (newspapers); (2) production occurs serially (construction); (3) the product is perishable (vegetables); or (4) the product cannot be stored or storage is expensive (electricity, natural gas). Temporal specificity means that delay or threats of delay by input suppliers or buyers can be very effective holdup strategies because of the difficulty in finding acceptable substitutes (input suppliers or buyers) on short notice.

Contracts

• The holdup problem suggests why firms might be reluctant to rely on spot markets to organize transactions when there are specific assets. But why can’t they use contracts to govern exchange? A contract is simply an agreement that defines the terms and conditions of exchange.

• Contracts align incentives by providing a mechanism for parties to a transaction to commit to their future behavior. If the implications of court sanction from nonperformance make a party worse off than performance, the incentive to act opportunistically by not living up to the terms of the contract will be attenuated, if not eliminated. And by incorporating contingencies, contracts allow for efficient adaptation. The contract can stipulate how the terms of exchange or trade will change as circumstances change.

Contractual governance and the holdup problem

• Changes in costs or demand change the potential total gains from trade, and efficient adaptation requires changing the terms of exchange to maximize the gains from trade given the new circumstances.

• This incentive to renegotiate could be tempered if the two parties agreed to a slightly more sophisticated contract. For example: cost-plus contract.

Complete vs. incomplete contracts (1)• A complete contract is one that will never need to be revised or chang

ed and is enforceable. It specifies precisely what each party is to do in every possible circumstance and for every circumstance the corresponding distribution of the gains from trade. And regardless of the circumstances a court will be able to enforce the contract-it is capable of requiring compliance and imposing damages such that both parties to the contract will honor the terms of the contract. This type of contract would provide no opportunities for renegotiation or holdup since it would contain no gaps, or missing provisions. However circumstances unfolded, the contract would unambiguously govern the exchange.

• The costs associated with negotiating, reaching, and enforcing agreements are called transaction costs. If transaction costs were zero, then all contracts would be complete and in such a world the Coase theorem tells us that agreements would be efficient and all gains from trade exhausted. However transaction costs are not zero.

• The Coase theorem states that in the absence of transaction costs all gains from trade should be exhausted regardless of the assignment of property rights.

Complete vs. incomplete contracts (2)• The costs associated with writing and enforcing complete contracts are:• 1. The costs of determining or anticipating all of the possible

contingencies (things that might happen) to which the terms of exchange should be responsive to ensure efficient adaptation.

• 2. The costs of reaching an agreement for each of the relevant contingencies.

• 3. The costs of writing the contract in sufficiently precise terms that the contract can be understood and interpreted as intended by a court. The lack of precision of language may preclude describing contingencies, actions, and rewards accurately. The resulting ambiguity means that multiple interpretations regarding responsibilities and performance are possible. This is especially likely to be a problem when specifying quality or future actions.

• 4. The costs of monitoring. Asymmetries of information mean parties to the contract will have to incur costs of monitoring to identify which contingency has been realized. One or both parties to the transaction may either have private information or engage in private actions that are unobservable or hidden from the other side and that the contract is contingent upon.

• 5. The costs of enforcement. In the event of a failure to perform, or breach of contract, costs will have to be incurred to enforce the contract.

Complete vs. incomplete contracts (3)

• The effect of these transaction costs is that contracts will be incomplete. This has two important implications:

• 1. There will be unforeseen contingencies or gaps in the contract. Things will happen for which the contract does not provide guidance on how the terms of exchange will be adapted.

• 2. The language of the contract will be sufficiently imprecise that for many foreseen contingencies courts will have difficulty in determining what the obligations of the contracting parties were and what constitutes adequate performance and what does not. It will be difficult to specify and measure performance.

• The more complex the transaction or the more uncertain the future, the greater the costs associated with writing a complete contract. We would therefore expect that the greater the complexity and uncertainty, the more incomplete the contract.

• When contracts are incomplete, incentives are aligned imperfectly and there is the possibility of being disadvantaged by self-interested, opportunistic behavior-being held up.


• In a world of incomplete contracts, the possibility of opportunistic behavior gives rise to the following inefficiencies:

• 1. Complex contracts. • In anticipation of potential holdups, firms will write more complex

contracts.• 2. Costs of renegotiation.• Incentives for holdup imply that firms are more likely to have to

renegotiate the terms of exchange. Again this will add to the costs of contracting, and delays due to renegotiation when there is temporal specificity may result in significant loses.

• 3. Resource costs to effect and prevent holdup.• Firms may expend resources to elicit concessions and their trading

partners may expend resources to prevent being held up. Productive resources are diverted to activities that have private value (redistribution of surplus), but not social value (nothing is produced).

• 4. Unrealized surplus.• Failure to renegotiate and realize efficient adaptation will result in

unrealized gains from trade.


• 5. Ex ante investments.• Firms are likely to incur additional expenditures and investments to

avoid being locked in to a single supplier. These kinds of investments reduce the dependency of a firm on a single supplier and increase its bargaining power ex post. This practice is called second-sourcing. It may mean a loss in economies of scale and hence a decrease in productive efficiency.

• 6. Underinvestment in specific assets.• Firms may reduce their investment in specific assets, thereby mitiga

ting their exposure to opportunistic behavior. Alternatively, they might substitute more general production methods for one using specific assets. However, these more general production technologies are likely less efficient. In both cases there is a reduction in gains from trade. The problem of underinvestment in specific assets arises because holdup eliminates residual claimancy status. A firm does not capture at the margin all of the gains created by its investment.

Example: underinvestment in specific assets and the holdup problem

• Consider a supplier of a single unit of an input. Suppose that the buyer agrees to pay p. let the cost of production for the supplier be C=c(e)+e where the effect of increases in e, investment in a specific asset, is to reduce the costs of production. This means that dc/de<0. the profits of the input supplier are π=p-c(e)-e. find the level of cost reducing effort if there is no possibility of holdup and if the seller anticipates that the buyer will be able to appropriate half of the seller’s quasi-rents.

• Solution. If the supplier is assured that she will receive p, then she will set e such that the marginal benefit to her equals its marginal cost or

• –dc(e*)/de=1. the marginal benefit of an extra dollar in investment is the reduction in costs. This is –dc(e*)/de. The seller’s quasi-rents q equal p-c(e). If the buyer is able to appropriate half, then the expected payment to the seller is ph=p-(p-c)/2=(p+c)/2 and her profits are πh=[p-c(e)]/2-e. the marginal benefit to the seller of another dollar of investment is now only –(1/2)(dc/de) and her optimal investment eh, assuming opportunistic behavior leads to an equal sharing of her quasi-rents, is defined by –dc(e*)/de=2. obviously that the effect of the holdup on the incentives for investment by the seller is the same as if the cost of investment were to double. As a result, the effect of the holdup is to reduce the investment by the seller. This happens because some of the marginal benefit created by an extra dollar of investment is transferred to the buyer and not captured by the seller: she is no longer a residual claimant.


• Klein (1996) has identified another important cost associated with using long-term contracts. Klein observes that long-term contracts may also be a source of holdup. While long-term contracts may alleviate the holdup problem, they may also create holdup problems. Long-term contracts with rigid provisions that turn out ex post to be incorrect can create windfall gains and losses. That is, long-term contracts can make it difficult to realize efficient adaptation because they define the status quo. If one party is doing very well under the terms of contract, then it will be reluctant to renegotiate-at lest not without preserving its windfall gains.

Vertical integration (1)• The use of spot markets to organize a transaction ensures efficient

adaptation and cost minimization. However, efficient adaptation will be problematic if there are relationship-specific assets due to the potential for holdup. Opportunistic behavior can be mitigated through the use of contracts, but only incompletely and only at a cost. If the costs of writing complicated contracts and the inefficiencies associated with incomplete contracts-especially underinvestment in specific assets-are relatively large, the firm may want to consider internalizing the transaction.

• Vertical integration has two dimensions. (1) it involves a change in the ownership of assets. (2) it also involves difference in governance. Independent contractors to employees of the downstream firms.

Vertical integration (2)• Ownership • The owner of an asset has the right ti determine the use and dispositi

on of the asset. • In a world of complete contracts, ownership is irrelevant since the us

e of the asset can be specified for all possible contingencies. In a world of incomplete contracts, however, ownership of an asset is important.

• Ownership is equivalent to the allocation of residual control rights.• It is the owner who has the power to determine the use of the asset w

hen there are contractual gaps or ambiguous contractual provisions.• Ownership of the assets of an input supplier eliminates the holdup pr

oblem by removing the second transactor. The independent input supplier that after integration becomes a supply division of the integrated firm cannot withhold the use of those assets or threaten to withhold the use of the assets in exchange for better terms of trade.

Vertical integration (3)• Governance • Vertical integration entails a change in governance.• Coase: the transaction costs associated with using the market arose f

rom (1) searching out trading partners and (2) negotiating the terms of trade. When the input requirements are ongoing, it may well be more efficient to substitute the authority of management for the price system. Instead of purchasing input requirements, the firm hires or employs factors of production and they (labor in particular) agree, within limits, to take directions from management. It is the replacement of the price system by the conscious coordination of management that defines a firm.

• Alchian and Demsetz challenged Coase’s view of the firm. The authority of an employer is no more and no less than the authority of a consumer. The authority of the employer arises because it can either sue or fire the employee. Likewise a customer unhappy with the performance of a supplier can also either sue or fire. To speak of managing, directing, or assigning workers to various tasks is a deceptive way of noting that the employer continually is involved in renegotiation of contracts on terms that must be acceptable to both parties.

Vertical integration (3)• Grossman and Hart: it is not obvious that integration should (1) mak

e any more information available; (2) make it easier to write and enforce contracts; or (3) make people less opportunistic. If this is true, then the effect of vertical integration is only to change the allocation of residual rights of control.

• Williamson and Masten: the nature of governance does change when a firm integrates with an input supplier. Distinctions in governance associated with differing organization forms arise because of differences in their status and treatment under the law. Differences in governance are possible for 2 reasons:

• (1) Differences in legal obligations. Employees have different obligations to their employer than independent contractors have to their customers. Employees are held to a higher standard than independent contractors to (a) obey directions; (b) disclose information; and (c) act in the interests of their employer.

Vertical integration (4)• (2) Differences in dispute resolution. Contractual disagreements bet

ween independent firms are typically resolved by resort to third-party mechanisms-either the courts or an independent arbitrator. Disagreements within a firm-regarding efficient adaptation and the distribution of surplus-are resolved by top management. Furthermore, there is considerably less potential for disputes inside the firm to be resolved by the courts. Dispute resolution related to holdup problem is likely to be much more efficient internally because (a) the less formal nature of the mechanism (court proceedings versus internal meetings) creates flexibility and lowers costs; (b) management is more likely to be informed with the background and expertise to understand and resolve the dispute efficiently; and (c) management should be able to acquire, and at lower cost, accurate information about exogenous changes and the actions of the parties to the transaction.

• Masten concludes: differences in the responsibilities, sanctions, and procedures applying to internal and market transactions thus seem to support the greater discretion and control and superior access to information generally associated with internal organization.

Complete contracts and team production (1)• Alchian and Demsetz: team production provides a rationale for the e

xistence and nature of firms. • Team production arises when the productivity of one factor of produc

tion depends on the presence and interaction with other factors of production.

• Factors of production are more productive when they are members of a team than when they are used on their own.

• Team production leads to difficulties measuring the contribution or effort of each team member’s contribution to output. Difficulties with monitoring the marginal product of each team member provide them with an incentive to shirk and free ride on the efforts of other team members.

• Alchian and Demsetz define a firm by the rights of its owner.• The owner has the right: (1) to be the residual claimant; (2) to monito

r and observe the other factors of production; (3) to be the central locus with which all the other factors of production contract-as opposed to contracting among themselves; (4) to change the factors of production utilized-in particular, to change team membership; (5) to sell these rights.

Complete contracts and team production (2)• The owner’s role as monitor and residual claimant arises due to problem

s with identifying the effort exerted by employees and the opportunity that this asymmetry of information provides for shirking. The other team members hire the owner to observe their behavior, measure their productivity and contribution to output, and determine appropriate compensation. The owner is provided with incentives to exert effort efficiently because of their residual claimancy.

• We can explore the relationship between ownership, monitoring, residual claimancy, team production, and shirking with a simple model.

• Individual: π(e)=b(e)-c(e); db(e*)/de=dc(e*)/de; e*-the efficient level of effort.

• Team work (2-person case): πi(ei,ej)=T(e)/2-c(ei); assume that e=ei+ej, T(e)=b(ei)+b(ej) and T(e) split by the 2 members equally, then (1/2)db(ei

p)/dei=dc(ei

p)/dei. That is to say, when the team member i exert a little more effort, she bears the full cost of the effort dc(ei)/dei, but receives only half of the benefit. The other half goes to the other member in the team. This leakage reduces the incentives for each member to exert effort and as a result both have insufficient incentives to exert effort. Since all members benefit from the extra effort but do not share in the cost, each has an incentive to undersupply effort or shirk.

Complete contracts and team production (3)• In order for there to be an incentive to form the team, the net income

of each member must be higher with the team. This can only be the case if the individuals are more productive as members of the team so that the gross benefits from team production are sufficiently great to offset the loss of residual claimancy and its effort on incentives. That is T(e)>b(ei)+b(ej)

• (1/2)dT(eiT*)/dei=dc(ei

T*)/dei, but the efficient level of effort should set the marginal benefit to the team equal to the marginal cost: dT(ei

T*)/dei

=dc(eiT*)/dei

• A solution to the shirking problem is to hire a monitor who tries to measure input and distributes output. This monitor helps to ensure that each member exerts the optimal effort level. Of course, the use of a monitor to mediate incentive problems is in and of itself costly: the monitor does not work for free. Provided payment to the monitor is less than the increase in output from increasing effort by the team members, the team members will be better off paying for a monitor whose task it is to stop them for shirking.

• But who monitors the monitor? To avoid the monitor having an incentive problem, the efficient response is for her to pay each of the members a fixed amount and in return the monitor becomes the residual claimant.as the residual claimant the monitor then has the correct incentives to exert the optimal amount of effort in monitoring.

Complete contracts and team production (4)

• The Alchian and Demsetz explanation does not explain why the difficulties associated with incomplete contracts are mitigated by vertical integration, and therefore it does not really provide an explanation for the extent of vertical integration. Economic organization does not matter in a world where the ability to contract is independent of its form. Economic organization will matter when

• (1) Contracts are incomplete and• (2) Contracting costs vary with the form of organi

zation.

Limits to the firm size (1)• Relationship-specific assets and the holdup problem strongly suggests

that spot market transactions are not always the optimal means to coordinate trade between input suppliers and their customers. The holdup problem can be mitigated through contracts, but only imperfectly, and contracts are costly. Vertical integration involves changes in ownership and governance, both of which suggest that internalizing transactions reduces transaction costs, ensures efficient adaptation, and improves incentives for investment. If this is true, why are there any market transaction at all? Why is not all production carried on in one big firm? What are the factors that limit the size of a firm?

• In the absence of relationship-specific assets, there are three advantages to using the market: (1) efficient adaptation; (2) cost efficiency; and (3) economies of scale. The existence of relationship-specific assets suggested an advantage for vertical integration on the basis of adaptation.

• The limit to firm size must be due therefore to cost disadvantages. These arise from not taking advantage of economies of scale and from incentive problems that lead to cost inefficiency.

Limits to the firm size (2)• Williamson observes that the cost disadvantages from not taking advant

age of economies of scale would not occur if the firm could sell its excess output to others. Because of contracting problem, others might not be willing to source supply of an input from a competitor. Of course the firm could merge with its competitor-expand horizontally-to solve this contracting problem.

• Merging with an input supplier results in a loss of high-powered incentives for the input supplier. The incentives of an independent supplier (residual claimant) to engage in innovation and cost minimization are likely greater than the incentives of a division.

• Incentive problems within the firm arise because of information asymmetries. There exist 2 kinds of informational asymmetries. (1) management may have better information about demand and costs than owners. (2) the actions of managers may not be perfectly observable. So managers have the opportunity to pursue their own objectives, which are not necessarily the objectives (cost minimization and profit maximization) of the firm’s owners. In particular, managers can exert suboptimal effort or direct resources of the firm toward uses that are not the firm’s interest, but provide them with consumption benefits. This type of behavior is referred to as managerial slack.

• The costs associated with (1) providing incentives, (2) monitoring managers, and (3) managerial slack are collectively referred to as agency costs.

The paradox of selective intervention (1)

• Suppose initially that A purchases an input from B. in response to potential holdup problems, A buys B and the owner of the input supplier becomes the manager of the new subsidiary or division. In order to preserve high-powered incentives (residual claimancy) and the advantages of vertical integration for efficient adaptation, the arrangement between the two divisions has the following 3 features:

• (1) A formula determines the price at which the input is transferred from division B to A. the determination of the transfer price might be the same as the contractual provisions between A and B when they where independent.

• (2) The income of the manager of the input supply division (its former owner) is the profits of the division. This makes the manager a residual claimant of the division and is suppose to preserve high-powered incentives.

• (3) The supply division will accede to requests by the firm to adapt efficiently to new circumstances. The firm will intervene between its two divisions only selectively to ensure efficient adaptation. Top management will intervene to ensure efficient adaptation between the two divisions of the firm, eliminating the holdup problem.

The paradox of selective intervention (2)

• The paradox arises by recognizing that just as contracts between independent firms are incomplete, so too are contracts within the firm. Consequently, holdup within the firm is possible and very tempting when incentives are high powered. In addition, managers will incur costs in an effort to redistribute gains or surplus within the firm-hold up other managers. This rent-seeking by management imposes so called influence costs on the firm.

Problems maintaining high-powered incentives• Williamson identifies asset utilization losses and accounting games as

the means to execute holdups internally.• 1. Asset utilization losses• Asset utilization losses arise if the firm has difficulty measuring the

economic profit of the supply division. The profits of the supply division can be conceptually divided into 2: (1) total revenues less variable costs and (2) changes in the value of the assets. Unlike revenues and variable costs, changes in the value of assets may be difficult for the firm to observe or measure. The value of the assets to the firm in the future will depend on usage and maintenance decisions made by the manager today. (Forgoing maintenance and extensive usage of capital goods)

• 2. Accounting games• The firm is in a position ex post to determine transfer prices and the

costs of production of the input division simply by changing accounting rules. Changes in the rules for determining the profits of the input division are possible due to contractual incompleteness.

• As a result, firms will neither be able nor find it desirable to maintain high-powered incentives. Rather they will find it advantageous to substitute low-powered incentives-make their managers salaried employees and subject them to administrative monitoring and controls. This substitution will result in a reduction in incentives for managers to exert effort, raising the costs of the firm.

Influence costs• A second resolution of the paradox of intervention is influence costs. The

ability to selectively intervene implies someone with the authority to make decisions and resolve disputes.

• Making good decisions, however, requires information, and the person with the authority to intervene is going to have to depend on others for most of the necessary information. And unfortunately, in many instances the source of information will be impacted by the decision. This provides employees with an incentive and opportunity to influence decision making. For instance, they can selectively and strategically present information to their advantage.

• The costs incurred by employees seeking to influence decision making are called influence costs. These are costs incurred by employees trying to influence those in authority to redistribute benefits in their favor.

• The absence of complete contracts internally coupled with asymmetric information means that employees can try and manipulate management to selectively intervene not to ensure efficient adaptation, but rather to execute a holdup-redistribute in the employee’s favor.

• Influence activities are costly to the firm for 2 reasons: (1) employees expend effort to influence those in authority-and counter the efforts of others-rather than pursuing the objectives of the firm; (2) to the extent that employees are successful, the firm’s decision are likely suboptimal.

Property rights approach to the theory of the firm (1)

• The view that contracting within a firm is just as difficult as contracting between firms is the starting point of the analysis of Grossman and Hart.

• Vertical integration does not change the nature of governance, but it does change ownership and therefore the allocation of residual rights of control.

• Allocation of residual rights control matters when contracts are incomplete because the holder of residual rights of control-the owner of the asset-determines the use of the asset when there are missing contractual provision.

• Ownership will affect the relative bargaining power over quasi-rents ex post. The owner of an asset will have greater bargaining power in a relationship because in the event of a breakdown in negotiations over surplus ex post, the owner gets to determine the use of assets.

• Different ownership structures will differentially affect the incentives of firms to make relationship-specific investment. Total gains from trade ex ante will often depend on relationship-specific investments. Incentives to invest depend on the ex post distribution of surplus and that depends on ownership. The benefit of integration is that the incentives of the acquiring firm to invest increase, but the costs of integration are a reduction in the incentives of the acquired firm to invest in relationship-specific assets.

Property rights approach to the theory of the firm (2)

• If the relationship-specific investment of the buyer (seller) is more important to creating gains surplus than the investment of the seller (buyer), then the buyer (seller) should own all of the assets-downstream (upstream).

• If the gains from trade depend on investments by both parties, then vertical separation-each transactor should own their own assets-is optimal, since it provides incentives for both parties to make investments.

• There are thus costs and benefits of integration and these costs and benefits are related to the effect that the allocation of residual rights of control (via ownership of non-human assets) has on the incentives for investment.

Grossman and Hart: an example (1)

• Suppose that the arrangements between an input supplier and a firm the production of the input B requires utilization of an asset and denote this asset as b. likewise production of output good A involves using an asset denoted a. moreover, suppose that the input supplier can exert effort to reduce costs. Let e represent the cost-reduction effort of the input supplier. This investment is relationship-specific. Similarly, the producer of A can make relationship-specific investments that increase the value of A. let I represent the dollar value of the investment by the buyer (firm A). In the first stage each firm makes its investment decision. In the second stage, the downstream firm would like to acquire a unit of the input from its supplier.

• The investment in I and e are noncontractible. This means either that the party that makes the investment cannot be compensated by the other or that if compensation is possible, it is not possible to verify that the investment was actually made. Moreover, it is assumed that the upstream (downstream) firm cannot make investment in i (e). The most natural interpretation, therefore, is that i and e are investments in human capital.

Grossman and Hart: an example (2)• It is also assumed that the price in the second stage is noncontractible: the

price of the input cannot be committed to via contract. In the second stage, the two parties will have to bargain over the terms of trade and if mutually acceptable terms cannot be reached, they will terminate their relationship. The simplest justification for this contractual incompleteness is that while the firm and the input supplier both know initially that demand will be for one unit of the input, they do not know until the second the kind of input required. At the beginning of the second stage this demand uncertainty is resolved and they learn the characteristics of the input required.

• We assume that the outcome of the negotiations is an efficient operating decision: the firms will come to an arrangement that maximizes the gains from trade (their profits) ex post. Since there will be relationship-specific investments, this means they will trade with each other and not pursue their next best alternatives. We assume that bargaining at the operating stage results in an equal division of the quasi-rents. Ownership determine the value of the outside options if there is no trade and hence affects the distribution of profits in the second stage.

• There are 3 possible ownership structure: (1) vertical separation-the input supplier owns asset b, the downstream firm asset a; (2) downstream integration-the input supplier acquires the manufacturer of A, that is input supplier owns both assets a and b; (3) upstream integration-the downstream firm acquires the input supplier

Efficient level of investment (1)

• The profits of the downstream firm when there is trade with its input supplier at price p for the input are: πA

e=v+2ai1/2-p-i. where v and a are both positive. The value of output when the downstream firm does not make any relationship-specific investment is v. the parameter a reflects the productivity of investments by the downstream firm. Increasing i increases the value of the downstream output and hence profits. The rate of increase in profits from an increase in i, or the marginal benefit of i, is: dπA

e/di=a/i1/2, which is positive, but decrease as I increases. Downstream profits are increasing, but at a decreasing rate, in the downstream firm’s investment.

Efficient level of investment (2)• The costs of the supplier when it trades with the downstream firm are:

CBe=s-2αe1/2. Where s is its costs in the absence of any relationship-sp

ecific investment and α>0 determines the productivity of that investment. Increases in investment or effort by the input supplier reduce the cost of production. The rate at which costs decline as effort increases, or the marginal benefit of e, is: dCB

e=- α/e1/2, which is negative and its absolute value decreases as e increases. Increases in e reduce costs, but the rate of decrease becomes smaller as e increases-there are declining marginal benefits in e. the profits of the input supplier when there is trade with the buyer and the transaction price is p are: πB

e=p-(s-2αe1/2)-e.

• The aggregate gains from trade between the two firms after the investments in i and e (stage 2) equal the revenues of the downstream firm less the avoidable costs of the upstream firm:

• V(2)=v+2ai1/2- (s-2αe1/2).• The ex ante (stage 1) aggregate gains from trade between the two firm

s are: V(1)=v+2ai1/2- (s-2αe1/2)-i-e.

Efficient level of investment (3)

• The efficient levels of relationship specific investment by the upstream and downstream firms maximizes the value of trade ex ante. The optimal value for i and e are found by setting their marginal benefit (MB) equal to their marginal cost. Since they are measured in dollars, the marginal cost of another unit of either i or e is one. The efficient I called i*, satisfies: MB(i*)=1 or, using dπA

e/di=a/i1/2 and simplifying: i*=a2.

• Similarly, the efficient level of investment by the input supplier is found by equating its marginal benefit to marginal cost. After simplification this becomes: e*=α2.

• If we substitute the efficient values for i and e back into V(1), we find that total profits when both firms make the efficient level of relationship-specific investment are: V*=k+ a2+ α2, where k=v-s is the value of trade when there is no relationship-specific investment.

Vertical separation (1)• Suppose now that there is no integration. The input supplier owns as

set b and the downstream firm asset a. in the event that they do not trade with each other, what are their outside alternatives? Suppose that the downstream firm can acquire the input from another supplier at price p0. however, if it does so its investment in i is less efficient and its profits in the second stage (after i has been made) are: πA

VS(2)=v+2ci1/2-p0, where c<a reflects the loss in value associated with investment in i from switching suppliers.

• On the other hand, the input supplier can also produce for another buyer and receive price p0. However, because e is relationship-specific its cost of production rises to: CB

VS=s-2γe1/2, where γ<α reflects the loss in value associated with investment in e from a switch in buyers. The profits of the input supplier in the second stage (after e has been made) are: πB

VS(2)=p0-(s-2γe1/2).• Ex post (after making the relationship-specific investments) total profi

ts for the two firms if they do not trade with each other, but instead exit, is: VVS(2)=k+2ci1/2+2γe1/2.

• Total profits if they do trade is: V(2)=k+2ai1/2+2αe1/2.

Vertical separation (2)• Since I and e are relationship-specific investments the value of trade i

s greater than the value of their outside options. The increase in profits from trading is the difference between V(2) and VVS(2), or

• Q=2(a-c)i1/2+2(α-γ)e1/2.• This of course equals available quasi-rents. Our assumption is that th

e input supplier and the firm realize V(2) because it is efficient-it maximizes aggregate profits-but divide the quasi-rents fifty-fifty. This means that the price p at which the input is traded between the two firms is determined by:

• v+2ci1/2-p0+(a-c)i1/2+(α-γ)e1/2=v+2ai1/2-p.• The left-hand side is the outside surplus of the downstream firm plus

half of the quasi-rents. The right-hand side is the surplus realized by the downstream firm if the two parties trade and the downstream firm pays p for the input. Solving for p, we find that the price paid for the input after bargaining will be: p= (a-c)i1/2- (α-γ)e1/2+p0.

• Based on this expected price under this ownership structure, the ex ante payoff-before its investment in i-for the downstream firm is

• πAVS(1)=v+2ai1/2-p-I

• =v+2ai1/2- (a-c)i1/2 +(α-γ)e1/2 -p0-I• =v+(a+c)i1/2 +(α-γ)e1/2 -p0-i.

Vertical separation (3)• The ex ante profits for the downstream firm under vertical separation reflect t

he ex post bargaining over quasi-rents. Instead of capturing the entire (or social) marginal benefit of its investment (a/i1/2), the downstream firm only expects to retain: dπA

VS(1)/di=(a+c)/2i1/2, since it must share equally the quasi-rents created by the relationship-specific aspect of the investment. The downstream firm is not a residual claimant with respect to its investment. Social marginal benefit is greater than the private marginal benefit since a>(a+c)/2 as a>c. The profit-maximizing choice of investment iVS is found by setting the downstream firm’s marginal benefit of investment equal to its marginal cost: (a+c)/2(iVS)1/2 =1, or, if we solve for iVS, iVS=(a+c)2/4.

• The ex ante profits of the input supplier under vertical separation are: πBVS

(1)=p-(s-2αe1/2)-e, or: πBVS(1)=(a+c)i1/2+(γ+α)e1/2+p0-s-e.

• The private marginal benefit of increasing investment in cost reduction for the input supplier is: dπB

VS/de(1)=(γ+α)/2e1/2.• If we set the private marginal benefit for the supplier (observe that this is aga

in less than the social marginal benefit because the downstream firm is able to capture some of the benefit as a result of ex post bargaining) equal to marginal cost, the optimal investment eVS in cost reduction by the input supplier when there is vertical separation is given by: eVS=(γ+α)2/4.

• The total profits under vertical separation are: VVS=k+(a+c)(3a-c)/4+ (γ+α) (3α- γ)/4.

Downstream integration (1)

Consider now the outcome if the assets of the downstream firm are acquired by the input supplier. The input supplier owns both asset a and b. we assume that the alternative income of the manager of the downstream firm is zero: in the absence of trade, the manager is fired. If there is trade between the two divisions, we assume that the net income of the manager is the profit of her division less her investment in human capital or effort. The upstream firm cannot either (1) make the investment in i or (2) compensate the downstream manager for her investment in i. In the non-trade case the integrated firm does not benefit from the expertise and human capital of the downstream manager and its profits are: πB

DS(2)=v-(s-2βe1/2), where β>γ>0 reflects that the productivity of investment by the input supplier is greater when it has access to both assets a and b. However α>γ: the full benefits of investment by the upstream firm require access to the downstream firm’s asset and its experienced manager.

If the manager of the downstream division is retained, then the aggregate profits of the integrated firm ex post are given by:

V(2)=v+2ai1/2- (s-2αe1/2).We assume that the owner of the integrated firm and the downstream man

ager are able to arrive at an efficient agreement with a 50:50 distribution of the quasi-rents.

Downstream integration (2)

• Taking the same steps as in the case of vertical separation, we can derive that the ex ante profits of the input supplier are: πB

DS(1)=v+ai1/2+(α+β)e1/2+p0-s-e, and the efficient level of investment by the input supplier is: eDS= (α+β)2/4.

• Similarly the ex ante income of the downstream manager (the profits of the downstream division less her investment in effort) will be: πA

DS(1)=ai1/2+(α-β)e1/2-i, and the efficient level of effort by the manager is: eDS=a2/4.

• Aggregate profits under this ownership structure are: VDS=k+3a2/4+(α+β)(3α-β)/4.

Upstream integration (1)• The downstream firm integrates backwards and purchases the asset

s of its input supplier. The downstream firm owns assets a and b. The input supplier owner now becomes a manager. Her incomes is normalized to be zero if there is no trade and she is released. If there is trade between the input supplier and the downstream division, her income is equal to the profits of the upstream division less her investment in effort or human capital.

• In the no trade case the profits of the integrated firm in the second stage are: πA

US(2)=v+2bi1/2-s, where b>c reflects that the productivity of investment in effort by the downstream firm is greater when it has access to both assets. However, a>b reflects that the productivity of investment in effort by the downstream firm is maximized when downstream firm has access to bth the manager and asset of its input division.

• If the manager of the upstream division is retained, then the aggregate profits of the integrated firm ex post are given by V(2)=v+2ai1/2- (s-2αe1/2). We assume that the owner of the integrated firm and the upstream manager are able to arrive at an efficient agreement with a 50:50 distribution of the quasi-rents.

Upstream integration (2)

• Taking the same steps as in the case of vertical separation, we can derive that the ex ante profits of the downstream division are: πA

US(1)=v-s+(a+b)i1/2

+αe1/2-i, and the efficient level of investment by the downstream firm is: iUS=(a+b)2/4.

• Similarly the ex ante income of the upstream manager (the profits of the upstream division less her investment in effort) will be: πB

US(1)=(a-b)i1/2+αe1/2-e, and the efficient level of effort by the manager is: eUS=α2/4.

• Aggregate profits under this ownership structure are: VUS=k+(a+b)(3a-b)/4+3 α2/4.

The optimal ownership structure (1)

• If the investments upstream and downstream are not relationship-specific, then α=β=γand a=b=c. In this circumstances, the aggregate profits under vertical separation and the investment levels are the same as the efficient outcome. The prediction of the analysis is that there should not be common ownership of the two assets a and b. with common ownership, the owner is able to hold up the other manager and hence her private return from investment is less than the social return, leading to underinvestment.

• If there is asset specificity, then α>β>γand a>b>c. under all ownership structures there is underinvestment. The extent of the underinvestment depends on the extent of exposure to opportunistic behavior and this varies with the ownership structure. The downstream (upstream) firm’s investment is the most when there is upstream (downstream) integration and the least when there is downstream (upstream) integration. The effect of ownership structure on incentives for investment is shown in Figure.

iDS iVS iUS i* eUS eVS eDS e*

MBDS MBVS MBUS MB* MB*MBDSMBVSMBUS

1 1

Investment and ownership

The optimal ownership structure (2)• The optimal ownership structure when there is

asset specificity is the one for which aggregate profits are greatest. This will depend on the importance of investment upstream versus investment downstream. If investment upstream is important, then it will be more important to protect the upstream firm from holdup and downstream integration will maximize aggregate profits. Alternatively if downstream investment is relatively more important, then upstream or backward integration will be the efficient ownership structure.

Implications of different ownership structure

Ownership structure

Downstream investment

Upstream investment

Aggregate profits

Efficiency a2 α2 k+ a2+ α2

Vertical separation

(a+c)2/4 (γ+α)2/4 k+(a+c)(3a-c)/4+(γ+α)(3α-γ)/4

Downstream integration

a2/4 (α+β)2/4 k+3a2/4+(α+β)(3α-β)/4

Upstream integration

(a+b)2/4 α2/4 k+(a+b)(3a-b)/4+3α2/4

The optimal ownership structure (3)• The importance of the investment is determined by its relative producti

vity. In comparing upstream and downstream integration, the optimal structure depends on the relative magnitudes of investment productivity and their interaction with ownership. The larger (smaller) a and b and the smaller (larger) αandβ, the more likely upstream (downstream) integration is efficient.

• Upstream integration is optimal the larger a and the greater difference between b and c.

• The larger a the more productive downstream integration, and the greater the difference between b and c, the greater the incentives provided for it under upstream integration since it is this difference that determines the reduction in exposure to the holdup problem from integrating versus vertical separation. On the other hand, the smaller αandγ, the less important upstream investment and hence the less costly it is to reduce the incentives for investment in it by transferring control of asset b downstream.

Optimal ownership structure

Dominant ownership structure

Boundary

VUS> VDS (2a-b)b >(2α-β) β

VUS> VVS (2a-b)b-(2a-c)c>(2α-γ) γ

VDS> VVS (2α-β) β -(2α-γ) γ > (2a-c)c

Do firms profit maximize• In many instances firms are managed by professional managers. Th

is separation of ownership and control suggests that profit maximization might not be the objective of a firm. While shareholders of the firm are interested in maximizing profits, the managers of the firm are likely interested in maximizing their utility. If managers are better informed than shareholders about profit opportunities or if the actions of management are unobservable to shareholders, then managers will have some latitude to pursue their own self-interest, or shirk, at the expense of profit maximization.

• The extent to which managers find it optimal to pursue theirown interests is limited by internal and external factors. Internally, managers are limited by monitoring and the use of incentive contracts. External factors that constrain the ability of managers to shirk are (1) managerial labor markets; (2) capital markets; (3) bankruptcy; and (4) competition in the product market.

Shareholder monitoring and incentive contracts (1)• The question of how the owners of a firm can induce the manager to

pursue the owner’s objectives rather than their own is an example of a principal-agent problem.

• Principal-agent problems arise when there are asymmetries of information due to either hidden information or hidden actions and when the preferences of the agent are not identical to those of the principal.

• If principals cannot observe or determine the behavior of their agents, there is hidden information. This allows for the possibility of moral hazard. The agent (manager) agrees to exert effort (to maximize profits) in exchange for a payment (salary) from the owners of the firm.

• If the owners of the firm cannot observe the effort of the agent, the agent has an incentive-to the extent that effort is costly to him-to reduce his effort.

• Profits depend not only on the effort of the manager, but also on exogenous shocks to either costs or demand that are also unobservable to the firm’s owners. So low profits do not signal low effort and high profits might be due to good luck rather than high effort.

• If the principal is not as well informed as her agent, the agent may be able to select alternatives that further his interests, as opposed to the interests of the principal.

Shareholder monitoring and incentive contracts (2)• Owners-the shareholders of firm-can mitigate, at least in part, the opportunities

for managers not to profit maximize by monitoring management and through the use of incentive contracts. The company’s board of directors are representatives of the shareholders and their job is to monitor management and approve major investments and policies. In doing so they have a legal obligation to shareholders to try and ensure profit maximization. A second way to align the incentives of managers with those of the firm’s owners is to give the managers a claim on the company’s profits. The closer that variations in the firm’s profits are matched by variations in the manager’s income, the more high-powered incentives. Perfect residual claimancy occurs when the manager has the sole claim on variations in the firm’s profits.

• However, while being a 100% residual claimant provides the manager with the right incentives to exert effort and make decisions, it exposes him to considerable risk. The profits of the firm depend not only on the effort of the manager, but also on exogenous cost and demand shocks. As a result the income stream of the manager will be variable and he will bear risk-which if he is risk averse will reduce his welfare. If the owners of the firm are risk neutral, then an efficient allocation of risk requires that the manager be fully insured-his income will be invariant to the profits of the firm.

• The optimal incentive contract trades off the incentives for effort and the efficient sharing of risk. In order to get the manager to exert effort, he will have to bear some risk. In order to get the manager to accept this risk, his expected income has to be higher, thereby raising the expected costs to the firm. Since it trades off incentives and risk sharing, the optimal contract will typically involve both insufficient incentives for effort and a suboptimal allocation of risk.

An optimal incentive contract with hidden actions (1)

• Suppose that the profits of the firm in the favorable, or good, state of the world are πG=36, but in the unfavorable, or bad, state of the world profits are πB=6. Whether the good or bad state is realized depends on the realization of either a demand or cost shock. For instance, the good state might occur if demand for the product turns out to be high and the bad state is realized if the demand for the product turns out to be low. The manager of the firm can either exert high (eh) or low effort (el). If he exert high effort, the probability of the good state (ph) is 2/3 and the probability of the bad state is 1/3. if he exert low effort then the probability of the good state (pl) is reduced to 1/3 and the probability of the bad state increases to 2/3. suppose that eh=2 and el=1. Let the utility function of the manager be u=y1/2-(e-1), where y is his income and e is his effort. The utility of the manager is increasing in his income, but decreasing in effort. The next best alternative for the manager provides him with a reservation utility u0=1.

• What is the full-information employment contract? What contract should the owner of the firm offer the manager if his effort is contractible-observable and verifiable in a court? In order to get the manager to accept the contract and exert the contracted level of effort, he must receive sufficient income so that he realizes at least his reservation utility. Since increasing the salary of the manager decreases the firm’s profit, the firm should pay him just enough to make him indifferent between exerting the contracted effort and not.

An optimal incentive contract with hidden actions (2)• The individual rationality constraint specifies the level of income that j

ust makes the manager indifferent between exerting the contracted effort or not working for the firm. If the firm wants to contract for high effort, the individual rationality constraint is u(yh,eh)=u0 or (yh)1/2-(eh-1)=1, where yh is the minimum salary that must be offered to elicit high effort. Setting eh=2, we can solve for yh and find that yh=4. similarly, if the firm wants the manager to exert low effort, it must offer a salary that makes the manager indifferent between exerting low effort and his reservation utility. This salary is yl=1.

• If the manager is paid to exert high effort and does so, then the expected profits of the firm are πh=phπG+(1-ph)πB-yh. If we substitute in the assumed values for ph, πG, πB and yh=4, then πh=22. If the manager is paid to exert low effort and does so, then the expected profits of the firm areπl=plπG+(1-pl)πB-yl. If we substitute in the assumed values for pl, πG, πB and yl=1, then πl=15.

• A profit-maximizing firm when effort is observable would offer the manager the following contract to maximize its profits: if e=eh=2, then y=4 and if e≠eh=2 then y=0. since the utility level of the manager if he exerts high effort will be 1, and only 0 if he exert low effort, this contract provides sufficient incentives for high effort and profit maximization.


• However, this contract is not incentive compatible if effort is unobservable. The manager has an incentive to promise to exert high effort, but in fact exerts low effort. Doing so increases his utility from 1 to 2 and reduces the expected profits of the firm to 12: u=y1/2-(el-1)=2 and π=plπG+(1-pl)πB-yh=12.

• The firm could proceed as if the agent is going to exert low effort and offer a contract of y=1. the agent would optimally choose to exert low effort and in doing so realize his reservation utility. The expected profits of the firm would be πl=15. However, the firm owner can do even better by offering an incentive contract.

• An incentive contract ties the pay of the manager to the profits of the firm. This exposes the manager to risk: if the good state is not realized, his salary will fall. This provides him with incentives to exert effort in order to minimize the probability of the bad state and maximize the probability of the good state. Since this imposes risk on the manager and he does not like risk, he will have to be compensated. His average or expected salary will be greater, which reduces the expected profits of the firm.


• An incentive contract will specify that the manager be paid yG if the firm’s profits are πG and yB if the firm’s profits are πB. The firm will choose yG and yB to maximize its expected profits subject to two constraints. The first is that the manager will voluntarily accept the incentive contract-it must be individually rational. This requires that ph(yG)1/2+(1-ph)(yB)1/2-(eh-1)≥u0, where the left-hand side is the manager’s expected utility from the incentive contract if he exert high effort. If we substitute in the values for eh, ph, and u0, this becomes 2(yG)1/2/3+(yB)1/2/3-1≥1.

• The second constraint is the incentive compatibility constraint. It requires that the manager find it in his interests to actually exert high effort. The incentive compatibility constraint is ph(yG)1/2+(1-ph)(yB)1/2-(eh-1)≥ pl(yG)1/2+(1-pl)(yB)1/2-(el-1), where the left-hand side is his expected utility from the incentive contract if he exerts high effort and the right-hand side is his expected utility from the incentive contract if he exerts low effort. If we substitute in the values for eh, ph, and pl, this becomes 2(yG)1/2/3+(yB)1/2/3-1≥ (yG)1/2/3+2 (yB)1/2/3.

• Maximizing expected profits will involve minimizing the expected payment to the agent. Therefore the optimal solution must involve satisfying both of above two as equalities. Solving these two equations we find that yG=9 and yB=0. The optimal contract when effort is unobservable is to pay yG=9 if πG is realized and yB=0 if πB is realized.


• Relative to the certain income of yh=4 when effort is observable, the optimal incentive contract tilts the manager’s compensation: it is significantly greater if the good state is realized and significantly worse if the bad state is realized.

• Under this contract the expected income of the manager is phyG+(1-ph)yB=6, which is considerably larger than the payment that must be paid to elicit high effort when it is observable. Consequently the expected profits of the firm are reduced to 20.

• A measure of the agency costs to the firm is the difference between its expected profits when effort is observable (the first best) and the optimal incentive contract (second best) when effort is not observable.

• The incentive contract may involve optimal effort but suboptimal risk allocation and may involve both suboptimal allocation of risk and suboptimal effort. It is also possible that high effort is optimal under full information, but cannot be induced with an incentive contract because it imposes too much risk-risk that requires simply too large an increase in expected salary relative to the expected increase in profits.

• The price of shares reflects the long-term prospects of the firm. Incentive contracts for managers that include stocks and/or stock options are superior to incentive scheme based only on profits or sales.

• The internal competition among managers to reach the top of the firm’s hierarchy and its reward of a relatively rich incentive contract reduces the incentives of managers not to cost minimize or profit maximize.

External limits to managerial discretion • Owners of a firm have two important rights:• (1) ownership of a share gives property rights in the

profits of the firm; and• (2) these residual claims can sold or transferred.• Shares are transferable residual claims. • The existence of tradable residual claims can promote

profit maximization. • The existence of tradable residual claims reduces the

latitude of management not to maximize profits and minimize costs through the creation of a market for corporate control and the managerial labor market.

• In addition bankruptcy constraints and competition in the product market can mitigate the divergence of interests between shareholders and management.

Managerial labor markets

• Stock markets create incentives to analyze firm performance and prospects, including the ability and plans of management, and this information is capitalized in the price of the firm’s shares. The price that the firm’s shares trade for reflects outside information on the firm and its management. Managers who are judged not to have adequately protected and advanced the interests of shareholders will be penalized in the market for managers through lower compensation and a reduction in the value of their human capital. Concerns for careers and reputations will encourage managers to exert effort to advance the interests of shareholders.

The market for corporate control: takeovers

• Capital markets also contribute to the discipline of management by creating a market for corporate control. The existence of shares provides an avenue for changes in ownership and changes in management. Inefficient or ineffective management is reflected in reductions in the price of shares. This provides a potential profit opportunity for investors or competing managers to take over the firm and replace existing management. More efficient management results in an increase in profits and in the share price of the firm. The market for corporate control provides an avenue to replace managers who are inefficient ones. Indeed, the threat of takeover and job loss-coupled with concerns over managerial reputation-provides some incentives for management to act more eficiently.

Bankruptcy constraints• A limit on the inefficiency of managers is the possibility of bankruptcy.

Bankruptcy occurs when the firm are not able to service its debt. This happens when it does not generate sufficient cash flow to repay its debt on schedule and make its interest payments. Bankruptcy, at the very least, will attract unwanted attention to the decisions and efforts of current management, if not lead to their dismissal-again with consequences for their future employment.

• Owners of a firm can provide incentives for efficiency by consciously increasing the debt load of the firm. Most obviously this increases the threat of bankruptcy and enhances incentives for efficient management. Less obviously, however, is the effect on the resources available to management. Jensen (1988) has highlighted the fact that debt service is not optional, and available to management. Jensen defines free cash flow as cash flow in excess of that required to fund all of a firm’s projects that have positive net present value when discounted at the relevant cost of capital. Such free cash flow must be paid out to shareholders if the firm is to be efficient and to maximize value for shareholders. One form of shirking arises when management does not pay out free cash flow, but instead uses it for projects of interest to the managers or dissipates it through higher costs. Shareholders can end up with the free cash flow by using debt in return for their stock and in doing so they reduce the free cash flow available to managers.

Product market competition (1)

• Adam Smith observed that monopoly, besides, is a great enemy to good management, which can never be universally established but in consequence of that free and universal competition which forces every body to have resources to it for the sake of self-defence. Smith’s point is that in competitive markets there is little or no scope for management to be inefficient.

• Increases in competition can discipline management through 2 channels. It can lead to an increase in information regarding the effort of management and it can directly discipline management by reducing the opportunities for slack. The information role of increases in competition works either through more efficient incentive contracts or the reputation effects of the managerial labor market.

Product market competition (2)• 1. Yardstick competition. The presence of competitors changes the abilit

y of shareholders to exercise control over management: it decreases the problems associated with the separation between control and ownership (Tirole 1988). Suppose that the relationship between the profits of the firm and the effort of managers is π= π(e,θ), where e is the effort of management and θis a random variable that affects either demand or costs in the industry. Both e and θare unobservable to shareholders. The optimal incentive contract will provide incentives for greater effort by imposing risk on the manager. This requires that risk-averse managers be paid higher expected wages.

• The ability of management to shirk will depend on the information shareholders have about θ. Relative to a monopoly situation, the existence of competitors will provide shareholders with additional information about θ. By looking at the profits of other firms in the industry, they will be able to infer something about the effort level of their managers since θis likely to have the same impact on all firms in an industry. If the profits of other firms are high, but the profits of their firm are low, they could conclude that θ was favorable, but their management did not exert very much effort. The presence of competitors reduces the amount by which managers can shirk, and hence reduces costs. This is a variant of yardstick competition. A monopolist will have higher costs because there is no yardstick to compare its profits with and thus managers will have more latitude to shirk.

Product market competition (3)

• Two determinants of profits are managerial effort and the exogenous shocks.

• Meyer and Vickers (1997) term the increase in efficiency from competition, or more accurately, the availability of comparative performance information, the insurance effect.

• 2. Reputation effects. Because the provision of additional information can allow owners in the managerial labor market to distinguish the effects of shocks from effort more effectively, there are enhanced incentives for managers to exert more effort in order to maintain or establish a good reputation for effectiveness.

• Increase in product market competition can also reduce agency costs or managerial slack by reducing the opportunity to slack. Increases in competition can make it more difficult for managers to reduce their effort when conditions in the industry are favorable.

• Nickell (1996) concludes that competition leads to a reduction in managerial slack (static inefficiency). Perhaps more significantly, the real value of competition is its effect on growth. Increases in competition are associated with higher growth rates in productivity-improved dynamic efficienct.

Summary(1)• The advantages from being large arise from economies of scale and ec

onomies of scope. Economies of scale and scope arise because of indivisibilities-it is not possible to scale inputs proportionately as output is reduced.

• Economic organization does not matter in a world where contracts are complete. Contracts would be complete if there were no transaction costs. Economic organization matters only when contracts are incomplete and transaction costs vary across the form of organization.

• The advantages of using spot markets to source inputs are (1) efficient adaptation; (2) cost minimization; and (3) realization of economies of scale. The ability to switch suppliers cost-lessly ensures efficient adaptation; cost minimization arises because an independent firm will be a residual claimant; economies of scale are realized because independent suppliers can aggregate demands.

• Relationship-specific investments, or asset specificity, create quasi-rents that are destroyed if firms switch input suppliers. The productivity advantages of relationship-specific investments create incentives for firms to form long-term relationships with their input suppliers. Alternative governance alternatives include spot markets, contracts, and vertical integration. The alternatives differ in the costs of achieving efficient adaptation-the realization of all the gains from trade.

Summary(2)• Incomplete contracts mean that firms that make relationship-specific

investments run the risk of having their quasi-rents expropriated. This is called the holdup problem and it gives rise to inefficiencies, in particular underinvestment in specific assets and failure to realize all the gains from trade (inefficient adaptation).

• Vertical integration of input supply (making instead of buying an input) implies differences in asset ownership and governance. These reduce or eliminate the possibility of holdup, thereby reducing transaction costs, promoting efficient adaptation, and improving incentives for investment.

• The limits of vertical integration or firm size arise because incentive problems in firms lead to cost inefficiency. This cost inefficiency due to managerial slack arises because of the loss of residual claimancy when an inputs upplier merges with a buyer. Residual claimancy cannot be maintained inside the firm because the holdup problem is not completely solved by integration. Top management will be unable to commit not to intervene and hold up managers of divisions. Nor will they be immune from rent seeking-behavior by employees that gives rise to influence costs and redistribution of income within the firm.

Summary(3)• If asset specificity is low and the potential for influence costs is high,

then the problems associated with internal production suggest that the transaction should be organized by the market. If asset specificity is high and/or the potential for influence costs is low, then it is more likely that the transaction will be organized internally.

• Asset ownership is equivalent to the allocation of residual rights of control. The holder of residual rights of control determines asset use when there are missing contractual provisions. The property rights approach to the firm predicts that the pattern of asset ownership (and hence vertical integration) will depend on the relative importance of providing incentives for noncontractible investments.

• Asymmetries of information (hidden actions and hidden information) and differences in preferences provide management with the opportunity and incentive to pursue their own objectives rather than profit maximization. Managerial discretion is limited by shareholder monitoring, incentive contracts, managerial reputation effects, the market for corporate control, bankruptcy constraints, and competition in the product market.

Date post:	26-Dec-2015
Category:	Documents
Upload:	gyles-horace-pope
View:	223 times
Download:	2 times

Industrial Organization Taught by Dr. Prof. Fang Qiyun Textbook: Industrial Organization: A...

Documents